[Federal Register Volume 70, Number 72 (Friday, April 15, 2005)]
[Rules and Regulations]
[Pages 19992-20015]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 05-6942]



[[Page 19991]]

-----------------------------------------------------------------------

Part II





Environmental Protection Agency





-----------------------------------------------------------------------



40 CFR Part 63



National Emission Standards for Coke Oven Batteries; Final Rule

  Federal Register / Vol. 70, No. 72 / Friday, April 15, 2005 / Rules 
and Regulations  

[[Page 19992]]



ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 63

[OAR-2003-0051; FRL-7895-8]
RIN 2060-AJ96


National Emission Standards for Coke Oven Batteries

AGENCY: Environmental Protection Agency (EPA).

ACTION: Final rule; amendments.

-----------------------------------------------------------------------

SUMMARY: On October 27, 1993 (58 FR 57898), pursuant to section 112 of 
the Clean Air Act (CAA), the EPA issued technology-based national 
emission standards to control hazardous air pollutants (HAP) emitted by 
coke oven batteries. This action amends the standards to address 
residual risks under section 112(f) and the 8-year review requirements 
of section 112(d)(6).

DATES: The final rule amendments will be effective on April 15, 2005. 
Existing sources will be required to comply with the final rule as 
amended on July 14, 2005. The incorporation by reference of certain 
publications listed in the final rule amendments is approved by the 
Director of the Federal Register as of April 15, 2005.

ADDRESSES: The EPA has established a docket for this action under 
Docket ID No. OAR-2003-0051. All documents in the docket are listed in 
the EDOCKET index at http://www.epa.gov/edocket. Although listed in the 
index, some information is not publicly available, i.e., confidential 
business information or other information whose disclosure is 
restricted by statute. Certain other information, such as copyrighted 
materials, is not placed on the Internet and will be publicly available 
only in hard copy form. Publicly available docket materials are 
available either electronically in EDOCKET or in hard copy form at the 
Air and Radiation Docket, Docket ID No. OAR-2003-0051, EPA/DC, EPA 
West, Room B102, 1301 Constitution Ave., NW., Washington, DC. The 
Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through 
Friday, excluding legal holidays. The telephone number for the Public 
Reading Room is (202) 566-1744, and the telephone number for the Air 
and Radiation Docket is (202) 566-1742.

FOR FURTHER INFORMATION CONTACT: Mr. Bob Schell, Emission Standards 
Division (C439-02), Office of Air Quality Planning and Standards, 
Environmental Protection Agency, Research Triangle Park, NC 27711, 
telephone number (919) 541-4116, e-mail address: [email protected].

SUPPLEMENTARY INFORMATION: Regulated Entities. Categories and entities 
potentially regulated by this action include:

------------------------------------------------------------------------
                                      NAICS       Examples of regulated
             Category                code\1\            entities
------------------------------------------------------------------------
Industry.........................       331111  Existing by-product coke
                                        324199   oven batteries subject
                                                 to emission limitations
                                                 in 40 CFR 63.302(a)(2)
                                                 and nonrecovery coke
                                                 oven batteries subject
                                                 to new source emission
                                                 limitations in 40 CFR
                                                 63.303(b). These
                                                 batteries are subject
                                                 to maximum achievable
                                                 control technology
                                                 (MACT) requirements and
                                                 are known as ``MACT
                                                 track'' batteries.
Federal government...............  ...........  Not affected.
State/local/tribal government....  ...........  Not affected.
------------------------------------------------------------------------
\1\ North American Industry Classification System.

    This table is not intended to be exhaustive, but rather provides a 
guide for readers regarding entities likely to be regulated by this 
action. To determine whether your facility is regulated by this action, 
you should examine the applicability criteria in Sec.  63.300 of the 
national emission standards for coke oven batteries. If you have any 
questions regarding the applicability of this action to a particular 
entity, consult the person listed in the preceding FOR FURTHER 
INFORMATION CONTACT section. Worldwide Web (WWW). In addition to being 
available in the docket, an electronic copy of today's final rule 
amendments will also be available on the Worldwide Web (WWW) through 
the Technology Transfer Network (TTN). Following the Administrator's 
signature, a copy of the final rule amendments will be placed on the 
TTN's policy and guidance page for newly proposed or promulgated rules 
at http://www.epa.gov/ttn/oarpg. The TTN provides information and 
technology exchange in various areas of air pollution control.
    Judicial Review. Under section 307(b)(1) of the CAA, judicial 
review of the final rule amendments is available only by filing a 
petition for review in the U.S. Court of Appeals for the District of 
Columbia Circuit by June 14, 2005. Under section 307(d)(7)(B) of the 
CAA, only an objection to the final rule amendments that was raised 
with reasonable specificity during the period for public comment can be 
raised during judicial review. Under section 307(b)(2) of the CAA, the 
requirements that are the subject of this document may not be 
challenged later in civil or criminal proceedings brought by the EPA to 
enforce these requirements.
    Outline. The information presented in this preamble is organized as 
follows:

I. Background
II. Summary of the Final Rule Amendments
    A. What Are the Affected Sources and Emissions Points?
    B. What Are the Requirements?
III. Response to Major Comments
    A. Comments on the Overall Risk Program and Policy
    B. Risk Comments Specific to Coke Ovens
    C. Comments on Section 112(d)(6) Review Policy
    D. Specific Comments on Section 112(d)(6) Review of Coke Ovens
IV. Statutory and Executive Order Reviews
    A. Executive Order 12866: Regulatory Planning and Review
    B. Paperwork Reduction Act
    C. Regulatory Flexibility Act
    D. Unfunded Mandates Reform Act
    E. Executive Order 13132: Federalism
    F. Executive Order 13175: Consultation and Coordination With 
Indian Tribal Governments
    G. Executive Order 13045: Protection of Children From 
Environmental Health and Safety Risks
    H. Executive Order 13211: Actions That Significantly Affect 
Energy Supply, Distribution, or Use
    I. National Technology Transfer Advancement Act
    J. Congressional Review Act

I. Background

    EPA promulgated national emission standards for charging, door 
leaks, and topside leaks from coke ovens batteries at 58 FR 57898, 
October 27, 1993 (40 CFR part 63, subpart L) under section 112(d) of 
the CAA. Section 112(f)(2) of the CAA requires EPA to determine for 
each section 112(d) source category if the promulgation of additional 
standards is required ``in order to provide an ample margin of safety 
to protect public health.'' We also have

[[Page 19993]]

discretion to impose a more stringent emissions standard to prevent 
adverse environmental effect if such action is justified in light of 
costs, energy, safety, and other relevant factors. On August 9, 2004 
(69 FR 48338), we proposed amendments to the national emission 
standards for coke oven batteries that included more stringent 
requirements for certain by-product coke oven batteries to address 
health risks remaining after implementation of the 1993 national 
emission standards. The proposed amendments also included provisions 
pursuant to the 8-year review requirements of CAA section 112(d)(6).
    In our proposal preamble, we presented the maximum individual risk 
(MIR) estimate for coke oven emissions from those emission points 
subject to the 1993 national emission standards. The MIR estimate was 
200 in a million (69 FR 48346). We also explained at proposal that, as 
required under the Benzene NESHAP \1\ decision framework (codified in 
section 112(f)(2)(A) and (B)), we considered the level of risk from the 
limits in the 1993 national emission standards (i.e., 200 in a million) 
to be acceptable after considering several factors (69 FR 48347-48350). 
These factors included the number of exposed people with cancer risk 
level estimates greater than 1 in a million (approximately 300,000 
people or 7 percent of the exposed population), the number of people 
for whom cancer risk levels are greater than 100 in a million (less 
than 10 people), the estimate of annual incidence of cancer (0.04), and 
the projected absence of adverse noncancer effects.\2\ Also considered 
in the evaluation in the proposal was the protective nature of many of 
the assumptions leading to these estimates of potential residual risk.
---------------------------------------------------------------------------

    \1\ National Emission Standard for Hazardous Air Pollutants 
(NESHAP): Benzene Emissions from Maleic Anhydride Plants, 
Ethylbenzene/Stryene Plants, Benzene Storage Vessels, Benzene 
Equipment Leaks, and Coke By-Product Recovery Plants (54 FR 38044, 
September 14, 1989).
    \2\ All estimates of population risk and estimated annual 
incidence in these final rule amendments are based on an upper-bound 
cancer unit risk estimate, a 70-year exposure duration, and our best 
estimates of exposure concentrations; cancer risk estimates using 
best estimates for exposure duration and unit cancer risk would 
yield lower risk estimates.
---------------------------------------------------------------------------

    Under section 112(o)(7) of the CAA, we are required to issue 
revised cancer guidelines prior to the promulgation of the first 
residual risk rule under section 112(f) (an implication being that we 
should consider these revisions in the various residual risk rules). 
Since our August 2004 proposal, we have issued revised cancer 
guidelines and also supplemental guidance which deal specifically with 
assessing the potential added susceptibility from early-life exposure 
to carcinogens. We have considered our decisions in these final rule 
amendments in light of the revised cancer guidelines and supplemental 
guidance. The supplemental guidance provides an approach for adjusting 
risk estimates to incorporate the potential for increased risk due to 
early-life exposures to chemicals that are thought to be carcinogenic 
by a mutagenic mode of action. For these chemicals, the supplemental 
guidance indicates that, in lieu of chemical-specific data on which age 
or life-stage specific risk estimates or potencies can be determined, 
default ``age dependent adjustment factors'' can be applied when 
assessing cancer risk for early-life exposures to chemicals which cause 
cancer through a mutagenic mode. In light of this guidance, EPA has 
evaluated the available scientific information associated with 
pollutants emitted by coke ovens and believes it is appropriate to 
apply the default factors in the risk assessment supporting today's 
final rule amendments. The chief HAP emitted by coke ovens, coke oven 
emissions, is specifically enumerated in CAA section 112(b)(1). Coke 
oven emissions are likely to cause cancer through a mutagenic mode of 
action. We base this conclusion on the data on coke oven emissions 
mutagenicity which has been summarized by EPA\3\ \4\ and the 
International Agency for Research on Cancer,\5\ and reported in 
numerous, more recent studies available in the peer-reviewed 
literature. The result of that determination is that our individual and 
population cancer risk estimates for lifetime exposures that begin at 
birth and extend through adulthood will increase from proposal by a 
factor of 1.6, \6\ a factor that considers the assumption of constant 
exposure over the 70-year exposure duration (birth to adulthood) we 
used in estimating individual and population risk. These further 
assumptions of increased cancer potency and birth to 70-year residence 
of the entire population in the area assessed were not part of the 
proposed rule amendments.
---------------------------------------------------------------------------

    \3\ Carcinogen Assessment of Coke Oven Emissions: Final Report. 
U.S. Environmental Protection Agency, Office of Health and 
Environmental Assessment. EPA-600/6-82-003F. February 1984.
    \4\ ``Coke Oven Emissions.'' U.S. Environmental Protection 
Agency. Integrated Risk Information System (IRIS). 1989. Available 
at: http://www.epa.gov/irissubst/0395.htm.
    \5\ IARC Monographs Supplement 7. International Agency for 
Research on Cancer. 1987, page 176. Available at: http://www-cie.iarc.fr/htdocs/monographs/suppl7/coke production.html.
    \6\ The ``Supplemental Guidance for Assessing Susceptibility 
from Early-Life Exposure to Carcinogens' recommends applying default 
adjustment factors to early life stage exposures to carcinogens 
acting through a mutagenic mode of action. The Supplemental Guidance 
recommends an integrative approach that can be used to assess total 
lifetime risk resulting from lifetime or less-than-lifetime exposure 
during a specific portion of a lifetime. The following adjustments 
represent the approach suggested in the Supplemental Guidance: (1) 
For exposures before 2 years of age (i.e., spanning a 2-year time 
interval from the first day of birth up until a child's second 
birthday), a 10-fold adjustment; (2) for exposures between 2 and 
less than 16 years of age (i.e., spanning a 14-year time interval 
from a child's second birthday up until their sixteenth birthday), a 
3-fold adjustment; and (3) for exposures after turning 16 years of 
age, no adjustment. In applying this factor to population risk, risk 
bins shown in appendix I of the risk assessment document were 
multiplied by 1.6, and the populations associated with those new 
risk bins were recounted depending on whether the bin risks were 
greater than 1 in a million, 10 in a million, or 100 in a million. 
The cancer incidence value was directly multiplied by the 1.6 
factor. The analysis and more detailed calculations may be found in 
the docket for this rulemaking.
---------------------------------------------------------------------------

    Based on the supplemental guidance, we have revised our risk 
estimates by applying the default adjustment factors to account for 
increased susceptibility that might occur due to exposures that occur 
from birth to 16 years of age. The increased risk due to consideration 
of the exposures assumed to occur from birth to 16 years of age 
(included in the 70-year total exposure duration) results in a revised 
upper-bound estimate. For the source category associated with the 1993 
national emission standards, the revised MIR estimate is 300 in a 
million. We have chosen to also apply the default adjustment to other 
analyses used to support the determination that the MIR of 200 in a 
million was acceptable. However, we acknowledge that more refined 
modeling of exposure would be necessary to adequately express the 
effect of early life susceptibility to overall estimates of population 
risk. For example, not all individuals are expected to be born in the 
area assessed. Nonetheless, after application of the default adjustment 
factor, our conclusions in the proposed rule amendments do not change 
and further refinement of the assessment was not warranted. The 
assumptions of exposure initiation (at birth for all) and cancer risk 
for coke oven emissions based on the application of the supplemental 
guidance would affect the number of exposed people with cancer risk 
levels greater than 1 in a million (500,000 people or 12 percent of the 
exposed population), the number of people exposed to risk levels 
greater than 100 in a million (approximately 70 people), the annual 
incidence of cancer (0.06), and the uncertainty associated

[[Page 19994]]

with the estimates of risk. The remaining factors we considered (e.g., 
actual emissions versus allowable emissions and the projected absence 
of adverse noncancer effects) are unaffected.
    Although we are adjusting risk estimates upward to reflect the new 
supplemental guidance, these estimated risk increases must also be 
tempered by consideration of other factors that were discussed at 
proposal and in the risk assessment document, and the further 
protective assumption added to the risk assessment that all individuals 
are born in the assessed area. For example, the coke oven battery 
sources are consistently controlling emissions below the level allowed 
by the 1993 national emission standards, which results in a 30 percent 
reduction in the estimated MIR. Our 70-year exposure assumption 
includes exposures from birth to 70 years. If exposures were from 3 
years to 73 years, the adjustment factor would be less than 1.6. If 
exposures were from 16 years to 86 years, no adjustment would be 
necessary. In addition, we used a health-protective assumption of a 70-
year exposure duration in our risk estimates; however, using the 
national average residency time of 12 years would reduce the estimate 
of risk by a factor of six (69 FR 48347). Our 1984 unit risk estimate 
(URE) for coke oven emissions is considered a plausible upper-bound 
estimate; actual potency is likely to be lower. After considering all 
of these factors, we continue to consider the MIR due to emissions at 
the limits in the 1993 national emission standards to be an acceptable 
level of risk (within the meaning of the Benzene NESHAP decision 
framework discussed at 69 FR 48339-48340, 48347-48348). As mentioned in 
the recently published cancer guidelines, we will continue to develop 
and present, to the extent practicable, an appropriate central estimate 
and appropriate lower and upper-bound estimates of cancer potency. 
Development of new methods or estimates is a process that will require 
independent peer review.
    We also re-examined our decision as to what level of control is 
necessary to provide an ample margin of safety to protect human health 
in light of applying the early-life exposure default adjustment 
factors. The 2010 lowest achievable emission rate (LAER) levels (which 
we are adopting as residual risk standards in today's action) will 
reduce the MIR from exposure to coke oven emissions to 270 in a 
million. In addition, the reductions will result in approximately 
200,000 fewer people having excess lifetime cancer risks of greater 
than 1 in a million from exposure to these emissions. After considering 
these estimates and the other factors explained in detail in the 
preamble to the proposed rule amendments, we continue to believe that 
the 2010 LAER levels provide an ample margin of safety to protect 
public health.
    The proposal allowed a 60-day comment period ending October 8, 
2004. The EPA's EDOCKET system logged a total of 16 public comments in 
Docket Number OAR-2003-0051. Commenters included one state association, 
two state agencies, a coalition of three major environmental groups, 9 
industry trade associations, one steel company, and two individual 
commenters. Each of their comments is summarized in our response to 
comments document contained in the rulemaking docket.

II. Summary of the Final Rule Amendments

A. What Are the Affected Sources and Emission Points?

    The affected sources are each coke oven battery subject to the 
emission limitations in 40 CFR 63.302 or 40 CFR 63.303 (the 1993 
national emission standards). There are five affected sources in this 
category: Four existing by-product recovery batteries and one 
nonrecovery battery. The final rule amendments apply to emissions from 
doors, topside port lids, offtake systems, and charging on existing by-
product coke oven batteries. Provisions are also included for emissions 
from doors on new and existing nonrecovery batteries and charging on 
new nonrecovery batteries.

B. What Are the Requirements?

    For existing by-product batteries, the final rule amendments limit 
visible emissions from coke oven doors to 4 percent leaking doors for 
tall batteries and for batteries owned or operated by a foundry coke 
producer. Short batteries are limited to 3.3 percent leaking doors. 
Visible emissions from other emission points are limited to 0.4 percent 
leaking topside port lids and 2.5 percent leaking offtake systems. No 
change has been made to the limit for charging--emissions must not 
exceed 12 seconds of visible emissions per charge. Each of these 
visible emission limits is based on a 30-day rolling average. The final 
rule amendments replace the less stringent limits that became effective 
on January 1, 2003, for MACT track batteries and are equivalent to the 
limits that will become effective on January 1, 2010, for batteries 
subject to LAER track requirements. We have not changed the standards 
for new by-product batteries.
    The monitoring, reporting, and recordkeeping requirements in the 
existing national emission standards continue to apply to existing by-
product coke oven batteries on the MACT track. These requirements 
include daily performance tests to determine compliance with the 
visible emission limits. Each performance test must be conducted by a 
visible emissions observer certified according to the test method 
requirements. A daily inspection of the collecting main for leaks is 
also required. Specific work practice standards must also be 
implemented if required by the provisions in 40 CFR 63.306(c). Under 
the existing standards, companies must make semiannual compliance 
certifications; report any uncontrolled venting episodes or startup, 
shutdown, or malfunction events; and keep records of information needed 
to demonstrate compliance.
    We are also issuing amendments for the improved control of charging 
emissions from a new nonrecovery battery (i.e., constructed or 
reconstructed on or after August 9, 2004). Fugitive charging emissions 
are subject to an opacity limit of 20 percent. A weekly performance 
test is required to determine the average opacity of five consecutive 
charges for each charging emissions capture system. The certified 
observer must determine and record the highest 3-minute average opacity 
for each charge; compliance is based on the average of the highest 3-
minute averages for five consecutive charges. Emissions of particulate 
matter (PM), a surrogate for particulate HAP in coke oven emissions, 
from a charging emissions control device are limited to 0.0081 pounds 
per ton (lb/ton) of dry coal charged. A performance test using EPA 
Method 5 (40 CFR part 60, appendix A) is required to demonstrate 
initial compliance with subsequent performance tests at least once 
during each title V permit term. If any visible emissions are observed 
from a charging emissions control device, the owner or operator is 
required to take corrective action and follow up with a visible 
emissions observation by EPA Method 9 (40 CFR part 60, appendix A) to 
ensure that the corrective action had been successful. Any Method 9 
observation of the charging emissions control device greater than 10 
percent opacity must be reported as a deviation in the semiannual 
compliance report. The final rule amendments also require the owner or 
operator to implement a work practice standard designed to ensure

[[Page 19995]]

that the draft on the oven is maximized during charging.
    We are also promulgating a work practice standard for the control 
of door leaks from all nonrecovery coke oven batteries on the MACT 
track. The owner or operator is required to observe each coke oven door 
after each charge and record the oven number of any door from which 
visible emissions occur. If a coke oven door leak is observed at any 
time during the coking cycle, the owner or operator must take 
corrective action and stop the leak within 15 minutes from the time the 
leak is first observed. After a door leak has been stopped, no 
additional leaks are allowed from doors on that oven for the remainder 
of that oven's coking cycle.
    We are allowing an exception to the 15-minute limit period for 
stopping a door leak. The owner or operator may have up to 45 minutes 
to stop a door leak no more than twice per battery during any 
semiannual reporting period. The limit of two occurrences does not 
apply if a worker must enter a cokeside shed to stop a leaking door 
under a cokeside shed. In that case, the owner or operator may have up 
to 45 minutes to take corrective action and stop the leak. The owner or 
operator also must operate the evacuation system and control device for 
the cokeside shed at all times that there is a leaking door under the 
cokeside shed.
    The owner or operator of a nonrecovery battery is also required to 
identify malfunctions that might cause a door to leak, establish 
preventative measures, and specify types of corrective actions for such 
events in its startup, shutdown, and malfunction plan. The final rule 
amendments also include recordkeeping and reporting requirements 
necessary to demonstrate initial and continuous compliance.
    We are also amending the provision in 40 CFR 63.303(a)(2) for 
existing nonrecovery batteries to state that the work practice standard 
for charging also applies to new nonrecovery batteries. These work 
practices are described in 40 CFR 63.306(b)(6).
    We are requiring that the owner or operator of existing by-product 
coke oven batteries on the MACT track comply by July 14, 2005. See CAA 
section 112(f)(4)(A), which states that existing sources must comply 
with section 112(f) residual risk standards within 90 days of the 
standard's effective date. We are also requiring that nonrecovery coke 
oven batteries on the MACT track comply by July 14, 2005 (or upon 
startup for a new nonrecovery battery for which construction commenced 
after August 9, 2004).
    The basis for the final rule amendments is set out in the preamble 
to the proposed rule amendments (69 FR 48338) unless otherwise 
explained in our responses to the major comments in this preamble. Our 
responses to all the comments are included in the docket.

III. Response to Major Comments

A. Comments on the Overall Risk Program and Policy

1. Ample Margin of Safety
    Comment: One commenter argued that CAA section 112(f)(2) makes 
clear that EPA's residual risk standards must reduce the lifetime risk 
to the single individual most exposed to emissions from any one of 
these sources to less than 1 in a million. In contrast, another 
commenter stated that EPA has properly construed the statute as 
establishing a trigger under which EPA must undertake a residual risk 
determination but not as establishing the level of risk reduction that 
must be achieved and further stated that EPA is not required to provide 
protection that achieves the 1 in a million excess cancer risk level.
    Response: The commenter's argument that the statute requires 
section 112(f) residual risk standards to reduce cancer risk to a most 
exposed individual to less than 1 in a million lacks a basis in the 
statutory text or in policy. Section 112(f)(2)(A) does indeed require 
us to promulgate standards if the ``lifetime excess cancer risk to the 
individual most exposed to emissions from a source in a category or 
subcategory'' is greater than 1 in a million. It does not establish 
what the level of the standard might be. See ``A Legislative History of 
the Clean Air Act Amendments of 1990,'' page 1789 (Conference Report), 
stating that ``[s]ection 112(f) contains a trigger for standards for 
non-threshold pollutants. * * *'' Rather, the level of the standard is 
to ``provide an ample margin of safety'' to protect public health. 
``Ample margin of safety'' is to be interpreted under the two-step 
formulation established by the Benzene NESHAP and CAA section 
112(f)(2)(B).
    Under that formulation, there is no single risk level establishing 
what constitutes an ample margin of safety (69 FR 48348). Rather, the 
Benzene NESHAP approach codified in section 112(f)(2) is deliberately 
flexible, requiring consideration of a range of factors (among them 
estimates of quantitative risk, incidence, and numbers of exposed 
persons within various risk ranges; scientific uncertainties; and 
weight of evidence) when determining acceptability of risk (the first 
step in the ample margin of safety determination) (54 FR 38045). 
Determination of ample margin of safety, the second step of the 
process, requires further consideration of these factors, plus 
consideration of technical feasibility, cost, economic impact, and 
other factors (54 FR 38046). As we stated in our ``Residual Risk Report 
to Congress'' \7\ issued under CAA section 112(f)(1), we do not 
consider the 1 in a million individual additional cancer risk level as 
a ``bright line'' mandated level of protection for establishing 
residual risk standards, but rather as a trigger point to evaluate 
whether additional reductions are necessary to provide an ample margin 
of safety to protect public health. This interpretation is supported by 
the interpretive language in the preamble to the Benzene NESHAP, which 
was incorporated by Congress in section 112(f)(2)(B).
---------------------------------------------------------------------------

    \7\ Residual Risk Report to Congress. U.S. Environmental 
Protection Agency, Office of Air Quality Planning and Standards. 
EPA-453/R-99-001. March 1999.
---------------------------------------------------------------------------

    We consequently believe that the commenter's bright line approach 
is not supported by the statute. Indeed, it is likely incorrect as a 
matter of law.\8\ In any event, EPA has concluded that the flexible 
approach to risk acceptability and ample margin of safety set forth in 
the Benzene NESHAP is desirable in light of the complex judgments EPA 
will make under section 112(f). The commenter's rigid approach lacks a 
basis in sound policy as well.
---------------------------------------------------------------------------

    \8\ It is true that the Senate version of CAA section 112(f) 
mandated elimination of lifetime risks of carcinogenic effects 
greater than 1 in 10 thousand to the individual in the population 
most exposed to emissions of a carcinogen. (See ``A Legislative 
History of the Clean Air Act Amendments of 1990,'' pages 7598 and 
8518.) However, this version of the legislation was not adopted. The 
EPA believes that the (rejected) Senate version of section 112(f) 
shows that Congress was capable of mandating a level of risk 
reduction had it wished to do so.
---------------------------------------------------------------------------

    Comment: Two commenters contended that EPA rejected a more 
stringent standard because the control technologies were not available 
at a reasonable cost. The commenters maintained that the more stringent 
standard would reduce risks to an acceptable level, and that the EPA 
does not have statutory authority to consider costs. According to one 
commenter, section 112(f) clearly calls for costs to be considered only 
in the area of adverse environmental effects.
    In contrast, a third commenter stated that EPA should not require 
any further reductions unless those reductions will produce discernible 
results stating that EPA justified the proposed additional reductions 
based on costs, yet noted that the reduction in cancer risk was so

[[Page 19996]]

small that it was within the noise level of EPA's ability to estimate. 
The commenter did not believe it was good policy to require additional 
reductions if EPA cannot be sure they will result in any benefit.
    Response: The first two commenters are mistaken regarding the 
consideration of costs in determining ``ample margin of safety.'' While 
it is correct that EPA does not consider costs in the first step (the 
``acceptability'' determination) of the ample margin of safety 
determination, costs are a factor which must be considered in the 
second step of the process (54 FR 38046).\9\ We have considered costs 
here in the authorized and required manner in assessing ample margin of 
safety after determining if baseline risk (level of risk remaining 
after imposition of MACT) is acceptable (54 FR 38045; 69 FR 48348-
48349).
---------------------------------------------------------------------------

    \9\ See also the Vinyl Chloride opinion at 824 F.2d 1146.
---------------------------------------------------------------------------

    In establishing an ample margin of safety, we weigh a range of 
factors, allowing flexibility on what constitutes an ample margin of 
safety (69 FR 48348). Some of the factors that can be considered are 
estimates of individual risk, incidence, numbers of exposed persons 
within various risk ranges, scientific uncertainties, weight of 
evidence, as well as potential standards' technical feasibility, cost, 
and economic impact. Balancing the above factors with the ability to 
achieve meaningful risk reductions is a critical component of the 
residual risk rulemaking process.
    We do not agree with the other commenter that the standards fail to 
produce discernible results. The emission limits are more stringent 
than the current MACT standards. The emissions reductions can be 
achieved at a nominal cost, they are technically feasible, and we 
estimate that the reductions will ensure that approximately 200,000 
fewer people having excess lifetime cancer risks of greater than 1 in a 
million.
2. Co-Located Sources and Facilitywide Risk
    Comment: One commenter said that many coke plants are part of a 
larger steel production complex; consequently, EPA should have 
considered the combined risk of all emission sources at the facility, 
including pushing, quenching, and battery stacks. The commenter also 
asserted that EPA should have considered the impact on residents near 
plants that are located in the same area (e.g., East Chicago and Gary, 
IN) and that the legislative history shows Congress' intent that EPA 
consider the combined risks of all sources of HAP emissions, regardless 
of source category, that are co-located. Specifically, Congress 
intended that the residual risk standards be stringent enough:

so that when all residual risk standards have been set, the public 
will be protected with an ample margin of safety from the combined 
emissions of all sources within a major source.\10\
---------------------------------------------------------------------------

    \10\ Floor Statement of Senator Durenberger in ``A Legislative 
History of the Clean Air Act Amendments of 1990'', vol. 1, page 868 
(Senate Debate on Conference Report).

The commenter disagreed with EPA's statement that delaying a full 
assessment of risk was a practical necessity because of the lack of 
information on actual emissions from pushing, quenching, and battery 
stacks. The commenter argued, essentially, that we are obligated to 
develop standards for the totality of risks simultaneously.
    Another commenter also stated that EPA should consider the facility 
as a whole and requested stringent controls on each source category to 
ensure the goals of the residual risk provisions are met in an 
expeditious manner. The commenter also asked that EPA ensure health 
protection in cases where there are multiple facilities in close 
proximity.
    Three commenters voiced opposition to consideration of emissions 
other than those from the specific source category at issue. One 
commenter indicated that the initial trigger for determining whether a 
residual risk standard was required at all must be applied only to a 
particular ``category or subcategory of sources'' (quoting CAA section 
112(f)(2)(A)). The commenter argued that the provision in section 
112(f)(2)(A) requiring us to develop residual risk standards if risks 
from the source category exceed a certain level also serves as a 
limitation in that ``residual risk determinations are to be done on a 
category or subcategory basis, not on a source or facilitywide basis.'' 
The commenter concluded that facilitywide risk could not be considered 
at all when establishing residual risk standards. According to this 
commenter, the only exception to a source category approach would be a 
voluntary request for a facilitywide determination so that they could 
use the most cost-effective set of reductions.
    Another commenter maintained that residual risk determinations for 
facilities as a whole would be acceptable only if EPA were to do so on 
a source category-by-source category basis. This commenter continued 
that if EPA were to adopt that approach, then the Agency cannot impose 
more risk reduction requirements on one source category to compensate 
for risks posed by another (co-located) source category.
    Another commenter argued that statutory language prevents 
consideration of risks posed by anything but the source category at 
issue, and further argued that any other approach would be difficult 
and confusing to implement. The commenter asserted that although EPA 
can consider facilitywide risk, residual risk standards should not be 
applied disproportionately to the first of the co-located sources 
evaluated in the residual risk process.
    Three commenters disagreed with EPA's use of Senator Durenburger's 
statement as the basis for the Agency's ``facilitywide'' interpretation 
of the statute. One commenter contended that the statement of one 
Senator cannot overcome the statutory language of section 112(f)(2) or 
the congressional directive to follow the Benzene NESHAP, particularly 
when the Senator noted that his remarks were not providing EPA specific 
new direction. Another commenter added that it was inappropriate to 
rely on the Senator's statements because the Conference Committee Joint 
Explanatory Statement suggests that the Senate and House Managers did 
not agree to much with respect to the Senate bill, and the Conference 
Report contains no explanation of section 112(f) on which EPA can rely 
for support.
    One commenter stated that a facilitywide approach would be bad 
policy because it would constrain the ample margin of safety for 
individual source categories beyond the level intended in the Benzene 
NESHAP framework. Trying to reconcile aggregated risk from dissimilar 
sources that may be geographically far apart may be difficult to 
accomplish and may not identify better opportunities for emission 
reductions (than would serial analyses for individual source 
categories). The commenter also stated that Congress directed EPA to 
establish a list of source categories and was well aware that many 
plants would have emission units falling into more than one category. 
Congress also anticipated that standards under section 112(d) and (f) 
would be staggered over time. The commenter contended that a 
facilitywide analysis could be too complex, speculative, and costly for 
other residual risk standards; therefore, EPA cannot and should not 
mandate facilitywide analyses in standards under section 112(f).
    Response: First, we should clarify the scope of the issue. Some 
discussion of

[[Page 19997]]

this issue has used loose terminology (i.e., ``facilitywide,'' ``co-
located,'' ``background'') as an imprecise shorthand for the various 
pollutant sources to which an individual could be exposed. In fact, 
there is a continuum of possible sources of exposure to consider. One 
could consider, in the initial assessment of residual risk from a 
source category, exposure from: (1) The individual emission points 
regulated under the standards being evaluated--here, charging, doors, 
lids, and offtakes--excluding all other sources, including nearby 
sources in the same category; (2) emissions from the source category 
only, but including co-located sources in the same category; (3) 
emission points at a facility that are necessarily co-located because 
they are part of an integrated common activity (e.g. pushing, 
quenching, and battery stacks for coke ovens); (4) all emissions at a 
facility (i.e., a stationary source or group of sources in any source 
category in a contiguous area under common control); (5) emissions from 
similar (or all) nearby facilities (``closely-located'' sources) whose 
emissions affect all or some of the same individuals; or (6) all 
ambient HAP, regardless of their source (e.g., automobiles, HAP 
originating from global sources).\11\
---------------------------------------------------------------------------

    \11\ Of course, in all of these cases, EPA would limit 
consideration to HAP emissions that are either the same as those 
emitted by the sources under evaluation or that have the same health 
effect or affect the same target organ.
---------------------------------------------------------------------------

    After considering the statute and the divergent views of commenters 
on these topics, EPA agrees with those commenters who stated that the 
natural reading of section 112(f) is that EPA should evaluate risks 
posed by the emissions only from the category or subcategory. Section 
112(f)(2)(A) instructs EPA to promulgate standards for ``each category 
or subcategory'' for which it has adopted MACT standards, if such 
standards are needed in order to provide an ample margin of safety to 
protect public health. The statutory ``trigger'' provision at the end 
of section 112(f)(2)(A), which mandates that EPA promulgate residual 
risk standards when ``cancer risks to the individual most exposed to 
emissions from a source in the category'' exceed a designated level, 
clearly is directed exclusively at emissions from the source category 
alone, and thus supports a reading that the ultimate requirement of the 
provision likewise applies only to emissions from the source 
category.\12\
---------------------------------------------------------------------------

    \12\ Further, section 112(c)(9) authorizes EPA to delist a 
category or subcategory on the basis of specified risk criteria. 
This section does not require EPA to look beyond the relevant 
category or subcategory in making delisting decisions. It would be 
inconsistent for Congress to allow categories or subcategories to be 
delisted entirely from the section 112 regulatory program using a 
category specific analysis, yet require EPA to look beyond the same 
specific category when making similar risk assessments under section 
112(f).
---------------------------------------------------------------------------

    We further agree, that while this is the first determination under 
section 112(f) since the adoption of the Clean Air Act Amendments of 
1990, Congress intended that EPA continue to apply the same test for 
determining when public health is protected with an ample margin of 
safety that was in effect before those amendments. Section 112(f)(2)(B) 
instructs EPA to use the ample margin of safety decision framework 
adopted in the Benzene NESHAP to make section 112(f) residual risk 
determinations, and indeed states that:

    [n]othing in subparagraph (A) or in any other provision of this 
section shall be construed as affecting, or applying to the 
Administrator's interpretation of this section, as * * * set forth 
in the Federal Register of September 14, 1989.

    In the Benzene NESHAP, EPA interpreted and applied the two-step 
test drawn from the D.C. Circuit's Vinyl Chloride opinion. Under that 
approach, EPA must first determine what level is ``safe'' ``based 
exclusively upon the Administrator's determination of the risk to 
health from a particular emission level.'' (See 54 FR 38055 (quoting 
Nat'l Res. Defense Council, Inc. v. EPA, 824 F.2d 1146 (D.C. Cir. 1987) 
(en banc)). The Court made clear, however, that ``safe'' does not mean 
``risk free.'' Id. Rather, the EPA must ``determine what inferences 
should be drawn from available scientific data and decide what risks 
are acceptable in the world in which we live.'' Id. In the second step 
under Vinyl Chloride and the Benzene NESHAP, once an ``acceptable 
risk'' level is determined, EPA must decide whether additional 
reductions are necessary to provide ``an ample margin of safety'' (54 
FR 38049). As part of this second decision, EPA may consider the costs 
of additional reductions, technological feasibility, uncertainties 
about available information or other relevant factors. Id.
    After examining the statutory scheme, the Benzene NESHAP, and sound 
policy concerns, EPA has concluded that, in its assessment of 
``acceptable risk'' for purposes of section 112(f), the agency will 
only consider the risk from emissions from that source category. This 
was the approach in the Benzene NESHAP, wherein EPA limited 
consideration of acceptability of risk to the specific sources under 
consideration (coke byproduct recovery plants, benzene storage vessels, 
benzene equipment leaks, ethylbenzene/styrene process vents, and maleic 
anhydride process vents) rather than to the accumulation of these and 
other sources of benzene emissions that may occur at an entire 
facility.\13\ See, e.g., 54 FR 38061 (stating in regard to 
consideration of natural background levels of a pollutant that 
``considering other sources of risk from benzene exposure and 
determining the acceptable risk level for all exposures to benzene, EPA 
considers this inappropriate because only the risks associated with the 
emissions under consideration are relevant to the regulation being 
established and, consequently, the decision being made.'') The Agency 
also rejected approaches that would have mandated consideration of 
background levels of benzene in assessing acceptability of risk. \14\
---------------------------------------------------------------------------

    \13\ EPA will consider, consistent with the Benzene NESHAP 
decision, whether co-location of entities within the same source 
category ``significantly influences the magnitude of the MIR or 
other risk levels'' (54 FR 38051). In this rulemaking, EPA has 
concluded that the health risks from the emissions at issue in this 
rulemaking are not affected (let alone significantly affected) by 
co-location with other entities in the same source category.
    \14\ EPA concluded that ``comparison of acceptable risk should 
not be associated with levels in polluted urban air'' (54 FR 38061). 
Background levels of certain HAPs can be relatively high, perhaps 
even above a level that might be considered ``safe.'' These 
background levels (including natural background) are not barred from 
EPA's analysis, but EPA will consider them along with other factors, 
such as cost and technical feasibility, in the second step of its 
112(f) analysis. To decide otherwise, EPA would have to conclude--
inconsistent with the Benzene NESHAP and sound policy--that 112(f) 
requires EPA to shut down any source that emits a HAP in an area 
with high background pollution, even if the emissions from that 
source are extremely small and do not appreciably affect overall 
risk.
---------------------------------------------------------------------------

    EPA has concluded that the sound policy embodied in the Benzene 
NESHAP remains the approach that EPA should follow in determinations 
under section 112(f). At the first step, when determining ``acceptable 
risk,'' EPA will consider public health risks that result from 
emissions from the source category only. Not only is this 
interpretation supported by the text of the statute and prior 
regulatory practice, but we are impressed and daunted at the practical 
problems of implementing a compulsory facilitywide examination. For 
example, as commenters pointed out, in future rules, the myriad 
combinations of source categories present at different facilities could 
create situations where nationwide consideration of residual risk 
becomes a practical impossibility because every facility would present 
a different fact pattern of source categories. Yet section 112(f) 
contemplates national determinations, not case-by-case evaluations and 
standards.

[[Page 19998]]

    At proposal, EPA cited a portion of a floor statement by Senator 
Durenberger as support for the position that EPA must assess the risk 
from an entire facility. EPA agrees with the commenters who stated that 
this statement is not sufficient evidence of Congressional intent to 
justify a different response than that adopted in the Benzene NESHAP, 
especially when, later in the same statement, the Senator states that 
section 112(f) is intended to be a ``return to current law'' under the 
Benzene NESHAP. (See Legislative History, Vol. 1 at 875-76.) As noted 
above, EPA did not adopt standards covering entire facilities in the 
Benzene NESHAP.
    This said, EPA disagrees that section 112(f) precludes EPA from 
considering emissions other than those from the source category or 
subcategory entirely. EPA must still determine whether additional 
reductions should be required to protect public health with ``an ample 
margin of safety.'' EPA believes one of the ``other relevant factors'' 
that may be considered in this second step is co-location of other 
emission sources that augment the identified risks from the source 
category. The Benzene NESHAP does not explicitly identify this as a 
relevant factor under step two, but the decision does acknowledge that 
``multiple exposures to chemicals are important to understand and 
consider in the EPA's overall implementation of its public health 
mandates' despite the fact that EPA has concluded that these risks 
should not be ``routinely evaluated and considered in selecting'' the 
level of acceptable risk (the first step of the Benzene analysis) (54 
FR 38059).
    The decision today is an example of a situation in which EPA has 
determined such a relevant factor merits evaluation. Each of the 
facilities subject to today's rulemaking is also subject to MACT 
emission standards on coke oven emissions from pushing, quenching, and 
battery stacks. These sources are necessarily co-located--they are 
integral parts of the same industrial activity. In this instance, EPA 
has the authority, in establishing ``an ample margin of safety,'' to 
impose greater reductions on a particular source category when the 
agency concludes that several of these co-located sources categories 
have elevated the overall public health risk to unacceptable 
levels.\15\ While this evaluation could be performed during the 
development of an individual residual risk standard for any particular 
source category that is part of a larger facility with multiple source 
categories, such an analysis would necessarily require sufficient data 
regarding the total facility emissions and the costs and risk impacts 
of reducing those emissions. Such information may conceivably be 
available when EPA does the first residual risk rule applicable to a 
facility, but it is much more likely that an early evaluation of cross-
category risks will be inconclusive due to a lack of complete 
information regarding other emission points. (In this rule, for 
example, EPA does not yet have an accurate quantification of pushing 
and quenching battery emissions reflecting these sources' operations 
under MACT standards; such information is needed to reasonably assess 
risks, costs, and further technologically feasible emission 
reductions.) EPA expects to develop better information about what cost-
effective emission and risk reduction opportunities are available as 
more source categories are assessed. EPA believes, in the future, it 
may be able to identify potential emission reduction trade-offs between 
co-located source categories that result in more efficient risk 
reductions for less economic cost at a facility.
---------------------------------------------------------------------------

    \15\ This is not to say that the EPA may impose significant 
reductions across an entire source category to alleviate health 
risks posed by co-location at a subset of facilities. In these 
circumstances, EPA believes it should further parse its emissions 
standards so as to impose greater reductions only on those 
facilities with significant co-location of other emissions. Put 
another way, EPA may permissibly develop section 112(f) standards 
that could result in different controls for co-located source 
categories at a facility than for the same source category which is 
not co-located.
---------------------------------------------------------------------------

3. Actual Versus Allowable Emission Rates
    We explained at proposal that we modeled emissions at the rates 
allowed by the 1993 national emission standards because they represent 
the source's potential emissions and risks and is, therefore, 
consistent with the language in CAA section 112(f)(2).
    Comment: We received some comments that agreed with the use of 
allowable rather than actual emission rates while other comments stated 
that we should use actual emissions. According to one commenter, 
Congress meant for EPA to make realistic estimates of residual risk. In 
support, the commenter pointed to the language of section 112(f)(2) 
which refers to a different measure of risk (i.e., risk to the 
``individual most exposed to emissions from a source'' rather than 
``maximum exposed individual'' or ``maximum individual risk'' used in 
the Benzene NESHAP) and associated passages in the legislative history. 
The commenter stated that EPA has data on actual emissions and should 
use this information as the basis for the risk assessment for coke 
ovens. Another commenter agreed with the decision to assume that 
sources are complying with the 1993 national emission standards when 
estimating emissions. The commenter also agreed with efforts to 
evaluate actual versus ``worst case'' potential emissions when 
estimating population risks and encouraged appropriate adjustments in 
future risk assessments. Another commenter stated that the use of 
maximum allowable emissions is particularly inappropriate for 
industrial source categories with batch operations because they 
consistently operate at levels well below the allowable rate.
    One commenter stated that EPA should not assume perfect compliance 
with allowable emission limits since several of these facilities are 
out of compliance. The commenter believed that we must account for 
noncompliance in the emission estimates.
    Response: EPA believes it may evaluate potential risk based on 
consideration of both actual and allowable emissions. This approach is 
both reasonable and consistent with the flexibility inherent in the 
Benzene NESHAP framework for assessing ample margin of safety. As a 
general matter, allowable emissions are the maximum level sources could 
actually emit and still comply with the national emission standards, so 
modeling this level of emissions is inherently reasonable for 
evaluating potential risks associated with current standards. As 
discussed in other sections of this preamble, coke oven battery sources 
are consistently controlling emissions below the level allowed by the 
1993 national emission standards, which results in a 30 percent 
reduction in the estimated MIR.
    It is also reasonable that we consider actual emissions, when 
available, as a factor in both steps of the determination (i.e., 
determining both risk acceptability and ample margin of safety). See 54 
FR 38047, 38050-38051, 38053 (we acknowledge a probable overestimate of 
emission levels in determining that risk and overall incidence is 
probably less than the maximum estimated levels). For the final rule 
amendments adopted today, years of monitoring data show that actual 
emissions have been consistently lower than allowable levels (69 FR 
48346-48347). Moreover, there is a sound empirical basis for coke oven 
emissions to be lower than theoretically allowable levels. To allow for 
process variability, sources typically strive to perform better than 
required by emission standards so that the emission

[[Page 19999]]

increases which occur on individual days due to process variability 
remain below emission standards. Failure to consider these data in risk 
estimates would unrealistically inflate risk levels.
    It is incorrect that a large number of these coke batteries are out 
of compliance. The batteries are inspected every day to determine 
compliance with the emission limits for doors, lids, offtakes and 
charging. We have compiled the results of these compliance inspections, 
and the details are in the rulemaking docket. The inspection results 
show that the coke batteries are operating consistently below the 
established emission limits and have shown essentially continuous 
compliance.
4. Exposure Duration
    Comment: Two commenters disagreed with the use of a 24-hour per day 
exposure over a 70-year lifetime to estimate individual and population 
cancer risks for refined risk assessments. According to one commenter, 
this exposure assumption is inconsistent with the recommendations by 
the National Research Council and the Commission on Risk Assessment and 
Risk Management. In their Reports to Congress, these organizations 
support development of distributional approaches to exposure 
characterization based on knowledge of the characteristics of a 
population's variability. This commenter asked EPA to develop a refined 
exposure methodology that incorporates information available on 
population residency times that will more accurately reflect population 
risk estimates. The development of this exposure methodology should 
also include a probabilistic analysis of estimated exposures. The other 
commenter stated that the use of such an unrealistic assumption makes 
the results overly conservative and will lead to additional and 
unnecessarily stringent standards more frequently than necessary.
    Response: We agree that our assumption that people may be present 
at their homes for 24 hours per day over a 70-year lifetime represents 
a scenario that likely overestimates the actual exposures received by 
people living near the facilities. Most people have daily activities 
that take them to areas where exposure concentrations are different and 
move to new residences periodically. Both of these behaviors will tend 
to lower lifetime exposures and, therefore, risk. The most significant 
risk reductions would occur for the group of people who are the most 
exposed. For these reasons, we are currently developing a methodology 
that will allow us to consider a variety of parameters (e.g., residency 
time, socio-economic conditions, age distribution, demographics, size 
of the census block) that could affect exposure and risk to individuals 
and populations that live in the vicinity of facilities. Other factors 
(e.g., emigration out of and immigration into the ``exposure area,'' 
social factors that affect population mobility, and census block size) 
may also influence the mobility of populations and, therefore, affect 
estimates of exposure and risk. As part of this effort, we are also 
investigating whether similar probabilistic techniques can be applied 
to the MIR to develop meaningful alternative metrics of individual 
risk. While this methodology is currently under development, we did not 
have sufficient information to apply any of these factors to these coke 
oven facilities.
    Finally, regarding recommendations of the Commission on Risk 
Assessment and Risk Management, we note that our overall approach is 
consistent with some of those recommendations. For example, the Risk 
Commission recommended that ``exposure assessments should not be based 
on a hypothetical MEI * * * should rely on more representative 
estimates or a maximally exposed actual person* * *.'' Our approach was 
based on identifying the maximum concentration where the census data 
identified people as actually living, and we assumed, as discussed 
above, that exposure of this individual was for 70 years starting at 
birth. Where we varied from the Commission's recommendation in this 
area was in assuming a 70-year exposure duration for the population as 
well. As just noted, we are developing a methodology that will allow us 
to look at the exposure variability that might be seen in the exposed 
populations. See the ``Residual Risk Report to Congress'' (at pages 
128-130) summarizing similarity in approaches.
5. Hazard Index
    Comment: Five commenters disagreed with use of the hazard index 
(HI) of 1 as the safe or acceptable level for noncancer health effects. 
One commenter stated that the HI level of 1 should be the ample margin 
of safety level because the values which form the basis for calculating 
HI already contain sufficient layers of safety to represent the ample 
margin of safety. The commenter contended that the reference 
concentration (RfC) or reference dose (RfD) represents the most 
stringent ample margin of safety level EPA should adopt.
    Three commenters recommended that EPA avoid establishing any bright 
line for a safe or acceptable level for non-carcinogens. One of these 
commenters explained that the HI of 1 would define both the acceptable 
risk level and the ample margin of safety level in one step, which is 
inconsistent with the two-step Benzene NESHAP framework. This commenter 
argued that an HI of 1 is too conservative because ``the ample margin 
of safety would always be set at or below an HI of 1.0, which would 
have an effect equivalent to a cancer level of 10-4 within 
the Benzene framework.'' The Commission on Risk Assessment and Risk 
Management's report selected a threshold HI of 10 because the RfC on 
which the HI is based already includes many uncertainty factors that 
should not be compounded in the ample margin of safety decision.
    Another commenter stated that EPA needs to clarify that the case-
by-case flexibility in the Benzene NESHAP framework also applies when 
interpreting hazard quotients (HQ) and HI. Although the proposal 
preamble did not identify a bright line, EPA's risk assessment document 
stated that an HI of 1 for each facility should ordinarily represent 
the safe or acceptable level, and that the ample margin of safety level 
may be lower or equal to the acceptable level, but can never be higher. 
The commenter objected because EPA was talking about an HI for a 
facilitywide analysis (rather than a specific source category) and 
because a rigid adherence to an HI of 1 for determining acceptable risk 
is unwarranted. The EPA should reserve flexibility in interpreting and 
applying HI and HQ acceptability, even in the screening stage. The 
flexibility is needed because of the variability in uncertainty 
factors, quality and consistency of data content, and other underlying 
information and assumptions. The commenter provided additional specific 
observations:
     In some cases, an HI or HQ can represent negligible or 
zero risk. There is no means to translate an HI or HQ into a 
probability of an individual incurring the effect (as is done for 
carcinogen effects).
     The EPA should do the initial screening using a target 
organ specific HI and should not aggregate across target organs and HAP 
for either the initial screening or refined assessment. No health-based 
conclusion can be reached from aggregating across different organs. An 
HI ``roll up'' for multiple chemicals' HQ must be predicated on target 
organ end points that are the same and a common mechanism or mode of 
action.

[[Page 20000]]

     Neither a range of 0.2 to 0.8 for HI nor a conservative 
default of 0.2 is permissible under the CAA. The statute only refers to 
the emissions and risk posed by a source category.
    Response: Five commenters pointed out that a statement in the risk 
assessment document indicated that an HI of 1 is the safe or acceptable 
level. Our statement in the risk assessment document was incorrect and 
has been revised. We did not use an HI of 1 as the acceptable level in 
our analysis. In the proposal preamble, we explained that ``the maximum 
estimated target organ specific HI for the emissions of HAP that may 
cause effects other than cancer from all emission points at the 
facility is 0.4,'' and that ``these emissions do not exceed a level 
which is adequate to protect public health with an ample margin of 
safety'' (69 FR 48350). Furthermore, we disagree that the ample margin 
of safety should never be more stringent (or less stringent) than the 
RfC (essentially an HQ or HI of 1) since, like the cancer framework, we 
do not consider an HI of 1 to be a bright line. We will evaluate the 
magnitude of the HI on a case-by-case basis.
    We disagree that an HI of 1 is equivalent to a cancer risk of 1 in 
10,000 as claimed by one commenter. As stated above, statements in the 
risk document identifying an HI of 1 as a safe or acceptable level are 
not correct and have been revised. We also disagree with the commenter 
who felt that the HI of 1 was too health protective because it did not 
consider different target organs. As used in the proposal and as 
intended for use in future residual risk assessments, the HI limit does 
reflect target organ specificity.
    The Commission on Risk Assessment and Risk Management's report does 
not say that an HI of 10 should be used as a level representing an 
ample margin of safety. The HI of 10 is used in that report in the 
context of screening (health-protective) risk assessments for residual 
risk. For sources with HI greater than 10, the Commission suggested an 
additional detailed risk assessment be performed. If the HI is still 
greater than 1, the facility is supposed to ``examine options/choose 
actions to reduce risk.'' For sources with HI between 1 and 10, 
facilities are supposed to voluntarily reduce emissions to achieve a 
lower risk category. The Commission recommended that if an HI is less 
than 1, no further action is required.
    We also note that most of these comments deal with conceptual 
issues not relevant to this rulemaking. We have not needed to make 
definitive determinations regarding appropriateness of any HI level 
because we have determined that exposures to emissions of threshold HAP 
from coke oven batteries (all emission points) are well within 
acceptable levels and require no further control to achieve an ample 
margin of safety.

B. Risk Comments Specific to Coke Ovens

1. Acceptable Risk
    Comment: Two commenters contended that EPA considered factors that 
might lessen the concern for risks, but did not give equal weight to 
factors that increase concern. For example, the EPA's analysis ignored 
HAP for which the Agency lacks cancer potency values.
    Response: We disagree with the commenters' concern that our 
analysis ignored HAP for which we lack cancer potency values. For those 
situations when cancer potency values are not in the Integrated Risk 
Information System (IRIS), we have established a prioritization process 
for accessing health assessment information from outside EPA (as 
described in our ``Residual Risk Report to Congress'' on pages 56 
through 58). This hierarchy includes dose-response values from EPA as 
well as other agencies that conduct scientific peer reviews such as the 
California Environmental Protection Agency Air Resources Board (CARB) 
and the Agency for Toxic Substances and Disease Registry (ATSDR), which 
is part of the U.S. Department of Health and Human Services. These non-
EPA values incorporate the best available science, are conceptually 
consistent with EPA's risk assessment guidelines, and have undergone a 
level of scientific peer review. Far from being ignored, many of the 
health assessment values used in the assessment were derived from non-
EPA sources (see Table B-1 in the risk assessment document).
    Comment: The risk is underestimated because EPA did not consider 
the risk from all carcinogenic HAP emitted from the facility.
    Response: As stated in the risk assessment document, inhalation 
cancer risk from the sources covered by the 1993 national emission 
standards was estimated using the HAP ``coke oven emissions,'' for 
which we have estimated a cancer URE. See CAA section 112(f)(6) which 
specifically acknowledges the possibility of considering risks of coke 
oven emissions as a whole; see also ``Residual Risk Report to 
Congress'' at page 108, noting that we may of necessity consider risks 
posed by the ``unique chemical substances'' enumerated in section 
112(f)(6), rather than attempting to ascertain every element of these 
complex mixtures and ascertaining a risk associated with each 
component. It is not necessary to consider separately the presence of 
each constituent of the mixture, coke oven emissions, which are also 
known to be carcinogens since their contribution to cancer risk is 
subsumed into the risk from the mixture. We considered the risk due to 
individual constituents when assessing non-inhalation and noncancer 
risks, when assessing risk from emission points where the composition 
of the mixture may be different (e.g., after the pushing emission 
control device), or when a screening level assessment was done. As 
described in the risk assessment document, we based our selection of 
HAP to be included in a screening level assessment on the availability 
of information on toxicity and emissions. Additional discussion of the 
HAP we considered is provided later in this preamble. The issue of HAP 
from co-located sources and facilitywide risk is discussed elsewhere in 
this preamble.
    Comment: One commenter stated that we should not accept a risk 
greater than 1 in 10,000 because of the weight of evidence that coke 
oven emissions, arsenic, and benzene are ``known'' human carcinogens. 
In support, the commenter cited the Benzene NESHAP * * * ``particular 
attention will also be accorded to the weight of evidence presented in 
the risk assessment of potential human carcinogenicity.''
    Response: While the commenter is correct that particular attention 
will be accorded to the weight of evidence presented in the risk 
assessment of potential human carcinogenicity, the weight of evidence 
is not the only health measure that must be considered. As stated in 
the Benzene NESHAP * * * ``no specific factor in isolation could be 
identified as defining acceptability under all circumstances'' (54 FR 
38044). Therefore, the acceptability of risk depends on consideration 
of a variety of factors and conditions. This assessment considered all 
of those factors listed in the Benzene NESHAP.
2. Ample Margin of Safety
    In the proposed rule, we said that even though emissions from 
pushing, quenching, and battery stacks are part of a different source 
category (because Congress singled out other emission points in section 
112(d)(8) and 112(i)(8)), they ``are an integral part of the same 
facilities covered by the national emission standards for charging, 
door leaks, and topside leaks (they not only are part of the same 
process but emit the same HAP)'' and

[[Page 20001]]

could permissibly be considered in setting the emission standard today 
(69 FR 48340). Table 1 of the proposed rule amendments (69 FR 48346) 
provided estimates of the risks posed by emissions from all components 
of the coking process at the four facilities (i.e. door, lid, offtake, 
charging, pushing, quenching, battery stack, and by-product plant 
emissions).
    As noted previously, EPA has not performed a complete residual risk 
determination for these other source categories, EPA has investigated 
the MIR and the population risk that result not only from the emissions 
being addressed by today's rulemaking but also from the other coke oven 
emission points located at the MACT track facilities. EPA's preliminary 
analysis has determined that emissions from the remaining coke oven 
facility emission points (pushing, quenching, battery stacks) do not 
cause risks appreciably greater in significance that those for the 
source category for which we are developing standards. Our risk 
estimates for pushing, quenching, and battery stacks are contained in 
the risk assessment document.
    EPA has concluded that delaying any further reduction is unlikely 
to result in disproportionate controls on other parts of a coke plant 
should EPA ultimately determine that further controls are necessary to 
provide an ample margin of safety. We therefore have determined that 
current information does not justify the imposition of more stringent 
controls to provide an ample margin of safety.
    Comment: One commenter suggested that EPA should also consider, in 
addition to the source category that is necessarily linked to the 
source category at issue, the risks from emissions from co-located iron 
and steel plants located within the same facility boundaries as the 
coking operations. Two of the four coke oven facilities affected by 
today's final rule amendments (AK Steel in Ashland, KY; and AK Steel in 
Middletown, OH) have integrated iron and steel plants co-located with 
their coking operations within their facility boundaries and under 
their control.
    Response: EPA does not believe it is appropriate to impose a 
restriction on all sources within a source category (here, the coke 
oven emission points at issue in this rule) based on the fact that half 
of the sources are co-located with a distinct source. The risk to 
public health from integrated iron and steel plants--sources which are 
not necessarily co-located with coke ovens--should be addressed in the 
residual risk determination for that source category. Nevertheless, EPA 
did assess the impact that emissions from co-located integrated iron 
and steel plants have on their facilitywide risk estimates. The 
integrated iron and steel plants are located fairly far from the coking 
operations at the two facilities where these two source categories are 
present at a common site. At Middletown, the iron and steel plant is 
located approximately 0.5 miles northeast of the coking operations. At 
Ashland, the iron and steel plant is located approximately 0.9 miles 
south of the coking operations. EPA's screening analysis indicates that 
the contribution of iron and steel emissions to the MIR posed by the 
coke oven sources is negligible.\16\ The MIR due to coking operations 
occurs to the west of the coking operation at the Middletown facility, 
and to the northwest of the coking operation at the Ashland facility. 
At both facilities the MIR is influenced by the proximity of the nearby 
population rather than by the primary wind direction, which is from the 
west/southwest. Stated simply, the iron and steel plants are located in 
such a way as to have only a very limited effect on those individuals 
who are most exposed to emissions from the coking operations. In fact, 
a reasonable rough estimate of the potential effect of integrated iron 
and steel plants on the MIR is less than 2 percent for both facilities.
---------------------------------------------------------------------------

    \16\ Even if a screening analysis suggested an important 
contribution from these sources, EPA would still need to consider 
more detailed assessments of sources and facilities with the highest 
risks. For example, in this screening analysis, EPA has treated iron 
and steel emissions as emanating from a single point (at a specific 
stack height). In a more detailed analysis, EPA would represent the 
actual plant configuration reflecting the disparate location of 
emission points and stack heights.
---------------------------------------------------------------------------

    Comment: Three commenters contended that the proposed amendments do 
not meet the requirements of section 112(f) or congressional intent 
because they do not protect the public health with an ample margin of 
safety. The proposed amendments would reduce risk from charging, doors, 
and topside leaks by only a small amount (from 200 in a million to 180 
in a million) and leave 200,000 people still exposed to risks greater 
than 1 in a million. One commenter said these risk estimates are ``in 
tension'' with EPA's general goals to protect the greatest number of 
people possible to a risk no higher than 1 in a million and to limit 
the risk to a person living near a plant to a risk no higher than 1 in 
10,000.
    Response: As noted earlier, we do not consider the 1 in a million 
MIR level as a ``bright line'' mandated level of protection for 
establishing residual risk standards. The final rule amendments will 
reduce the excess lifetime cancer risks for an estimated additional 
200,000 people to less than 1 in a million, a goal that is not ``in 
tension'' with our general goal of protecting the greatest number of 
people possible to risks no higher than 1 in a million. In determining 
the ample margin of safety (i.e., the level of the standard), health 
risk is one factor that we must consider, along with other factors such 
as cost and technological feasibility. Balancing these and other 
factors with the ability to achieve meaningful risk reduction benefits 
is a critical component of the residual risk rulemaking process. We 
considered reducing risks further but concluded that the technology 
required would be cost prohibitive for this industry and therefore 
undesirable.
3. Scope of the Risk Analysis
    Comment: The EPA's proposal did not contain any information on if 
or how the agency assessed the risks from acute exposure to coke oven 
emissions or how the proposed standards would protect public health 
with an ample margin of safety from such risks. The EPA ignored the 
recommendation from one peer reviewer on the need to justify no 
consideration of the health effects from acute exposure.
    Response: Risks from acute exposure are of greatest concern when 
excess emissions occur and cause a peak or spike in ambient 
concentrations of a pollutant. Coking is a continuous operation (i.e., 
the coke oven battery is operated continuously and is seldom shut down, 
other than for a major rebuild or extensive repairs, because the 
cooling during shutdown could damage oven walls). The ovens in a 
battery are in various stages of operation such that any emission 
fluctuations would be caught in the highly buoyant plume which rises 
continually above the batteries. From a toxicological perspective, 
references values derived for acute exposure assessment are higher 
concentrations than chronic reference values. Consequently, for 
situations, such as this, where there are not short periods of higher 
exposure levels, the chronic assessment will be controlling. In this 
assessment, no significant chronic non-cancer effects were identified, 
therefore, no acute effects would be expected.
    Comment: The EPA must assess exposure through eating food in which 
toxics have accumulated or bioaccumulated, drinking contaminated water, 
and dermal exposure through contaminated soil. And, while EPA 
considered fish consumption at recreational levels, it did not consider

[[Page 20002]]

risks to subsistence fishing population, including those on the Great 
Lakes and poor people in urban areas. Mercury, dioxins, lead, and PAH 
are examples of other toxics released from coke ovens whose primary 
risks are from non-inhalation pathways. The EPA must reassess the risk 
and include dietary pathways from all of the relevant pollutants. 
Another commenter recommended that EPA improve its multipathway risk 
assessment methods.
    The commenter stated that EPA admitted that its generic 
environmental analysis was not intended to be used to predict specific 
types of effects to individuals, species, populations, or communities 
or to the structure and function of the ecosystem. According to one 
commenter, EPA's failure to consider any impact on any individual 
species contravenes the CAA. Another commenter recommended that EPA 
develop criteria for refined ecological assessments that meet the 
statutory specifications.
    Response: The multipathway assessment used for this analysis was 
based on the multipathway assessment initially used for a secondary 
lead smelters case study and was refined through the use of EPA's most 
current multipathway guidance. These include, for example, EPA's Office 
of Solid Waste's peer-reviewed ``Human Health Risk Assessment Protocol 
for Hazardous Waste Combustion Facilities'' \17\ which provided overall 
guidance and chemical-specific values for bioaccumulative and 
persistent HAP.
---------------------------------------------------------------------------

    \17\ Human Health Risk Assessment Protocol for Hazardous Waste 
Combustion Facilities, Vol. 1 (peer review draft), U.S. 
Environmental Protection Agency, Office of Solic Waste and Emergency 
Response. EPA 530-D-98-001A. 1998.
---------------------------------------------------------------------------

    The HAP included in the analysis were selected using the procedures 
described in the risk assessment document and parallels the selection 
methodology described in our recently released ``Air Toxics Risk 
Assessment Reference Library.'' \18\ Additionally, we only included the 
HAP for which we had sufficient information to suggest that the HAP 
were emitted from the sources which are the focus of these final rule 
amendments and for which emissions could be estimated. The air toxics 
included in this assessment were a group of PAH and lead. The final 
rule amendments will reduce the amount of these emissions from coke 
ovens. Mercury would ordinarily be included in the list of persistent, 
bioaccumulative, and toxic (PBT) HAP to be assessed, but as discussed 
in the risk assessment document and in section III.B.4 of this 
preamble, mercury emissions were very low for this source category, 
primarily because volatile compounds like mercury are captured and 
removed in the by-product recovery plant.
---------------------------------------------------------------------------

    \18\ Air Toxics Risk Assessment Reference Library. U.S. 
Environmental Protection Agency, Office of Air Quality Planning and 
Standards. Vol. I: Technical Resource Manual, EPA 453-K-04-001A. 
Vol. II: Facility-Specific Assessment. EPA 453-K-04-001B. April 
2004.
---------------------------------------------------------------------------

    Multiple routes of exposure were assessed in the multipathway 
assessment including both inhalation and ingestion of contaminated 
food, soil, and drinking water. A mixture of best-estimate central 
tendency and health-protective assumptions were used in order to be 
health-protective for both adults and children, but also to estimate 
risks that were not beyond the level of plausibility. This assessment 
uses a ``farmer/recreational fisher'' scenario. In the scenario, the 
farmer/recreational fisher was located at the point of the maximum 
impact to agricultural land near each of the facilities, and our 
assessment included the consumption of all types of home-produced 
fruit, vegetables, beef, pork, and dairy products, as well as locally-
caught fish. The pathways included in this assessment were inhalation, 
soil ingestion, produce ingestion, fish ingestion, drinking water 
ingestion, and breast milk ingestion for infants. The farmer was 
assumed to consume locally-caught fish at the rate of a recreational 
fisher, but both central-tendency and high-end consumption rates based 
on values from the ``Exposure Factors Handbook'' were included in the 
analysis to increase confidence that individuals that may have higher 
consumption would be protected. Risks were estimated using the health-
protective assumption of lifetime continuous exposures.
    The screening-level ecological risk assessment used for this 
analysis used the same methods as the secondary lead smelters case 
study to estimate HAP media concentrations and to develop protective 
screening-level ecological toxicological dose-response values. This 
screening-level assessment was designed to identify and further 
evaluate HAP that pose a potential ecological risk and to remove from 
the analysis those HAP that did not pose such risks. In order to feel 
confident that this assessment considered threatened and endangered 
species, this analysis intentionally used assumptions that, overall, 
tend to overestimate risks. These assumptions include the following:
    Choice of ecologic receptor. This assessment evaluated the species 
from a broader list of species (sediment dwellers, including aquatic 
sediment dwellers), soil dwellers, aquatic life, air and soil dwelling 
plants, various representative types of mammals; see risk assessment 
document, Table 3-8) that are considered widely distributed and provide 
a representative range of body sizes and diets. In cases where multiple 
species from which to choose were available for a particular exposure 
scenario (e.g., a terrestrial herbivore), EPA evaluated the species 
with the lowest benchmark (i.e., the most sensitive species) for this 
assessment.
    Choice of risk metric. All species in the assessment are evaluated 
against the No Observable Adverse Effect Level (NOAEL). As the name 
indicates, this is a level of exposure below which one would not expect 
to see any adverse effects. Since relatively few animal or plant 
studies have determined these safe levels of exposure over an entire 
lifetime or several generations, a NOAEL for chronic exposures to a 
particular chemical must be estimated from toxicity studies of the same 
chemical conducted on a different species of wildlife or on laboratory 
animals. In these cases, to ensure that species survival is accounted 
for and to be more health-protective, whenever possible we used the 
NOAEL from studies in which more sensitive endpoints such as 
reproductive and developmental toxicity and reduced survival were the 
outcome as opposed to direct mortality. To evaluate potential risk to 
aquatic life, we used as a comparison benchmark EPA's Water Quality 
Criteria (adopted pursuant to section 304(a) of the Clean Water Act) 
which are used by States (and authorized Tribes) in adopting water 
quality standards for the protection of human health, aquatic life, and 
aquatic-dependent wildlife.
    Further protective assumptions related to exposure. We made the 
additional protective assumption that terrestrial and aquatic species 
reside and therefore forage and drink exclusively in the area where the 
maximum HAP concentration is estimated. We further assumed that any HAP 
to which they are exposed is 100 percent bioavailable.
    Protective assumptions related to emission levels. The ambient 
concentrations estimated for each terrestrial wildlife exposure 
scenario were derived from the modeling done for the human health 
assessment, and so contains the same protective assumption that 
emissions are constantly at the level allowed under the 1993 national 
emission standards. We know that actual emissions are less (69 FR 
48496-48497) and, therefore, exposure and risk would also be less.

[[Page 20003]]

We also assumed that the emissions from the coke facility with the 
highest emissions were representative of the emissions that might be 
anticipated from the other coke facilities subject to these final rule 
amendments.
    Even using these highly protective assumptions, modeled 
concentrations remain under the NOAEL for each species, in most 
instances by many orders of magnitude. For risks to aquatic life, 
modeled risks for each HAP again remained an order of magnitude lower 
than the Water Quality Criteria levels.
    We recognize that there are data limitations for these analyses 
that indicate a need for further refinement and development of 
multipathway and ecological risk assessment tools. The multipathway and 
ecological reference methodology described in the ``Air Toxics Risk 
Assessment Reference Library'' (see footnote 18) will be revised. While 
these more complex tools were not needed in the coke oven residual risk 
assessment (because no screening-level ecological effects were seen 
even when the assessment included many protective assumptions), they 
are important and may play a larger role in future residual risk 
assessments, and we will be developing future guidance.
    Comment: One commenter said that because HAP emitted by coke oven 
batteries is persistent and bioaccumulative, EPA was obliged to consult 
with the Fish and Wildlife Service as required by the Endangered 
Species Act. The commenter further stated that such consultation should 
consider information in EPA's Great Waters Report,\19\ issued pursuant 
to CAA section 112(m), that species are affected by deposition of HAP 
emitted by sources located in areas near the Great Lakes.
---------------------------------------------------------------------------

    \19\ Deposition of Air Pollutants to the Great Waters: Third 
Report to Congress. U.S. Environmental Protection Agency, Office of 
Air Quality Planning and Standards. EPA-453/R-00-005. June 2000.
---------------------------------------------------------------------------

    Response: Given the many protective assumptions of this assessment, 
we remain confident that if an individual member of a species is 
protected, as shown in our assessment, then the population as a whole 
would be protected. EPA has not identified any evidence of effect on 
critical habitat, given that our analysis shows no adverse effect on 
the terrestrial or aquatic life evaluated. Since our results showed no 
screening-level ecological effects, we do not believe that there is an 
effect on threatened or endangered species or on their critical habitat 
within the meaning of 50 CFR 402.14(a). Because of these results, EPA 
concluded a consultation with the Fish and Wildlife Service is not 
necessary. In this regard, we again reviewed the Great Waters Report 
mentioned in the public comment. There is no mention of threatened or 
endangered species in our ``Great Waters Reports to Congress.'' The 
risk assessment conducted in this rulemaking is consistent with the 
recommendations in the report to conduct assessments of the potential 
impacts of the emissions and deposition of PBT HAP on ecological 
systems, including water bodies.
    Comment: One commenter disagreed that there is no information that 
would allow EPA to assess the risk to children from coke oven 
emissions. All of the individual constituents in coke oven emissions 
have been studied in children, and children have been found to be more 
susceptible than adults to each of the toxic components. The commenter 
provided extensive information on why children's airways are more 
susceptible to airborne carcinogens and provided health effects 
information on PM, PAH, and mercury. The commenter stated that an 
adequate risk assessment must include the acute and chronic respiratory 
effects of PM; cancer, reproductive, and developmental effects of PAH; 
and the neurotoxic effects of mercury on children.
    Response: The commenter is mistaken; we did not state in the 
proposal preamble or risk assessment document that we had no 
information to assess the risk to children. We acknowledge that 
population subgroups, including children, may have the potential for 
risk greater than the general population due to greater body burden 
and/or greater susceptibility to the toxicant. Our risk assessment 
accounts for these greater body burdens. For certain exposures (e.g., 
lead), children were explicitly assessed, while in other cases (e.g., 
inhalation pathway) lifetime (rather than simply childhood) exposure 
was assumed, which would tend to yield higher estimates of risks.
    In the ingestion pathway assessment, risks to children from lead, a 
pollutant with known hazard to children from the ingestion pathway, 
were explicitly assessed and presented. As part of the multipathway 
screening analysis (see appendix A of the risk assessment document), 
blood lead concentrations were predicted for estimates of cumulative 
lead exposure of children aged less than or equal to 7 years old. As 
described in the risk assessment document, the predicted blood lead 
concentrations all fell below the Center for Disease Control level of 
10 micrograms per deciliter ([mu]g/dL), an indicator of elevated blood 
concentration. The maximum level estimated was 0.1 [mu]g/dL.
    While risks to children from other pollutants were not separately 
assessed for the ingestion pathway (only central tendency and high end 
adult values were estimated), we do not consider the ingestion pathway 
to be the driver or highest risk pathway. The amount by which exposure 
factors generally increase the resultant cancer risk of children (less 
than 18 years of age) over a similar exposure duration for adults is 
less than a factor of three. Review of the ingestion pathway cancer 
risk estimates for the adult exposures indicates that ingestion pathway 
cancer risk estimates for a similar duration of children's exposure 
would still fall below the inhalation pathway cancer risks. Given that 
the highest cumulative HI for the adult exposures was on the order of 
0.001, a separate estimate for children's ingestion exposure while 
expected to be a slightly higher value, would still fall well below an 
HI of concern. Consequently, the major focus for the risk assessment 
was placed on the inhalation analysis.
    In the inhalation pathway assessment, the exposure assessment 
described the maximum exposure of residents near coke oven emissions. 
The exposed population was presumed to be exposed to airborne 
concentrations at their residence continuously 24 hours per day for a 
full lifetime. No greater inhalation exposure to neighboring residents 
would be feasible.
    With regard to children's potentially greater susceptibility to the 
toxicants present in coke oven emissions, the assessment relied on 
Agency dose-response values which have been developed for all subgroups 
of the general population, including children. For example, a recent 
review \20\ of the chronic reference value process concluded that the 
Agency's RfC and RfD derivation processes adequately considered 
potential susceptibility of different subgroups with specific 
consideration of children, such that the resultant RfC/RfD values 
pertain to the full human population ``including sensitive subgroups,'' 
a phrase which is inclusive of childhood.
---------------------------------------------------------------------------

    \20\ A Review of the Reference Dose and Reference Concentration 
Process. U.S. Environmental Protection Agency. Risk Assessment 
Forum. EPA/630/P-02/002F. December 2002.
---------------------------------------------------------------------------

    With regard to cancer dose-response values, our revised cancer 
guidelines and new supplemental guidance recommend applying default 
adjustment factors to account for exposures occurring during early-life 
exposure to

[[Page 20004]]

those chemicals thought to cause cancer via a mutagenic mode of action. 
The effect of these guidelines on the risk assessment is discussed in 
detail in section I of this preamble.
    In summary, our dose-response values have been developed via 
methodology that is intended to provide either a plausible upper-bound 
potency factor or an exposure with which there is likely no appreciable 
risk of adverse effects during a lifetime considering all population 
subgroups, including children.
    Comment: One commenter asked EPA to faithfully apply the standards 
for ``influential scientific risk assessment information'' to the risk 
assessments that underlie residual risk rules. The commenter also asked 
EPA to implement and fully adhere to the Agency's Information Quality 
Guidelines so that the data and analysis will be sound and well 
represented to decision makers and the public. The commenter stated 
that EPA should aggressively pursue reform of its risk assessment 
practices in response to the advice of its key advisors, should take 
steps to eliminate conservative assumptions embedded in its risk 
estimation procedures, and should begin work on a recommended 
alternative approach that will produce more accurate and realistic 
estimates.
    Response: In compliance with the Agency's Information Quality 
Guidelines, specifically as they apply to influential scientific risk 
assessments, we have taken significant steps to ensure that the 
substance of the information in our risk assessments supporting the 
coke ovens residual risk rule is accurate, reliable, and unbiased. To 
this end, we have used the best available science and supporting 
studies as well as data collected by the best available methods. For 
example, many of the components of our risk assessments (air quality 
and exposure models, toxicity values, methods for estimating emissions, 
etc.) have undergone independent scientific peer review on their own or 
as applied in specific case studies. In addition, we have subjected the 
final report on the coke ovens risk assessments to a peer review by 
experts external to the Agency through a letter review process 
administered by a third party. Through this peer review, we have 
endeavored to ensure that the presentation of information on human 
health and environmental risks is comprehensive, informative, and 
understandable. The final risk assessment document, revised per the 
peer review, as well as the peer reviewers' comments and our responses 
to them, have been made available to the public in the docket for this 
rulemaking.
    Comment: Two commenters stated that the risk assessment was 
inconsistent with the Agency's Information Quality Guidelines because 
EPA did not use newer, peer-reviewed health effects data (i.e., using 
the 1984 IRIS value for coke oven emissions instead of newer, peer-
reviewed health effects data submitted by Sciences International).
    Response: The commenters pointed to a single study \21\ which 
interpreted only a portion of the health effects data available on coke 
oven emissions and was subjected to a scientific journal peer review. 
While such a study would not ordinarily be considered comprehensive 
enough or broadly-vetted enough to serve as a sole basis for risk 
estimates in this type of assessment (and indeed to do so could raise 
Data Quality Guideline issues), we did address the use of the 
alternately-derived cancer potency in our risk assessment (i.e., 
compared risk estimates reported in the IRIS and the newer values). 
Since the use of this value did not substantially affect the level of 
estimated risks or the associated risk-based decision, EPA undertook no 
further evaluation of these health effects data. In the future, 
however, newer assessments of health effects can be readily considered 
in the residual risk program if they are sufficiently comprehensive and 
vetted through an appropriate scientific peer review process.
---------------------------------------------------------------------------

    \21\ Moolgavkar, S., et al. ``Estimation of Unit Risk for Coke 
Oven Emissions.'' Risk Analysis, vol. 18, no. 6, pages 813-825. 
1998.
---------------------------------------------------------------------------

    Comment: One commenter said the risk assessment was inconsistent 
with the Agency's Information Quality Guidelines because EPA did not 
provide central tendency estimates (i.e., results were restricted 
primarily to upper bound estimates).
    Response: As pointed out by the commenter, we addressed the central 
tendency requirements of the Information Quality Guidelines in a 
limited way in the risk assessment that supports this rule. As noted 
above, the upper-bound potency value that is presented in IRIS is 
routinely characterized using the standard descriptor for the cancer 
potency (``upper bound''), by saying that the upper bound is not likely 
to underestimate risks, that true risks are likely to be less, and 
that, for some individuals, risk may be zero. As described in the 
Information Quality Guidelines and reiterated in the recently published 
cancer guidelines, we will continue to develop and present to the 
extent practicable an appropriate central estimate and appropriate 
lower and upper-bound estimates of cancer potency. Development of new 
methods or estimates is a process that will require independent peer 
review.
    We also understand that most people have daily activities that take 
them to areas where exposure concentrations are different and move to 
new residences periodically. Both of these behaviors may tend to lower 
lifetime exposures to coke oven emissions (i.e., lower than our current 
assumption of 70-year exposure duration), and therefore lower 
individual risk attributable to coke ovens. In the proposal preamble 
(69 FR 48347), we presented an alternative estimate of an individual 
risk level adjusted to reflect the national average residency time of 
12 years for comparison with the results from our 70-year exposure 
assumption. This change in assumption would result in a lowering of 
risk by approximately six-fold. It is important to note that if the 
cancer dose-response is reasonably linear with dose at environmental 
exposure levels, estimated individual risk attributable coke oven 
emission is lower for those living fewer years in the affected area, 
but estimates of total population incidence are not affected if the 
overall population remains stable (assuming people moving out are 
replaced by people moving in). Taking this into consideration and to 
provide better metrics by which to assess population risks in the 
future, we are currently developing a methodology that may allow us to 
consider a variety of parameters that could affect risk to populations, 
not just to the individual, that live in the vicinity of facilities. 
Other factors (e.g., emigration out of and immigration into the 
``exposure area,'' social factors that affect population mobility, and 
census block size) may also influence the mobility of populations and 
therefore, affect estimates of exposure and risk. As part of this 
effort, we are also investigating whether similar probabilistic 
techniques can be applied to the MIR to develop meaningful alternative 
metrics of individual risk. While this methodology is currently under 
development with subsequent evaluation and peer review to follow, we 
did not have sufficient information to apply any of these factors to 
these coke oven facilities.
    Comment: One commenter said that the risk assessment was 
inconsistent with the Agency's Information Quality Guidelines because 
EPA's reasoning for not conducting a more complete probability analysis 
was not sufficient.
    Response: We stated in our proposal that we ``considered the needs 
and

[[Page 20005]]

scope of the assessment'' before deciding whether to do a more refined 
population analysis and concluded that this ``level of refinement was 
not necessary * * * because the results of a probabilistic analysis are 
unlikely to affect the proposed risk management decisions.'' Our 
decision was that risks to the population at the level of the standard 
we proposed met the required ample margin of safety determination. 
Refining the population risk distribution by considering factors such 
as population mobility in the analysis would not change that decision, 
only refine the underlying results on which that determination was 
made. Therefore, we did not believe that the additional expenditure of 
time and resources to do that analysis was warranted. Also, in making 
this decision, we believe we are meeting the requirements of the 
Information Quality Guidelines by providing information that is 
accurate, clear, complete, and unbiased.
4. Mercury Standards
    Comment: One commenter contended that EPA's proposal was unlawful 
because it excluded controls for mercury. The commenter argued that EPA 
is required to establish emission standards for each HAP and that 
section 112(f)(2) requires EPA to consider every HAP that a category 
emits to ensure that the residual risk standards adequately protect 
public health and the environment. The commenter cited 2002 Toxic 
Release Inventory (TRI) data that show AK Steel (Ashland, KY) emits 27 
pounds of mercury and that Indiana Harbor Coke reported 650 pounds of 
mercury emissions.
    Another commenter questioned why mercury and other metals were 
excluded from door leak emission estimates. According to the commenter, 
mercury is highly volatile and would be expected to occur in emissions 
or leaks from any part of the process. The commenter also requested 
that EPA explain why mercury is missing from the list of metals that 
were monitored in appendix C of the risk assessment document. While 
mercury is listed as a component of coke oven emissions in one table in 
appendix C, it is unclear if or how EPA used this mercury emission 
factor in its analyses.
    Response: Our research indicates that most of the mercury that is 
volatilized from the coal during the coking process at by-product coke 
batteries is concentrated in the tar when the gas is processed in the 
by-product recovery plant.\22\ The vast majority of the volatiles 
distilled from the coal are collected and processed to recover by-
products. However, the commenter is correct in that emission tests have 
detected mercury emissions from coke ovens. For example, small 
quantities of coke oven gas may escape through leaks on doors, lids, 
and offtakes. The emission factor for mercury in Table C-23 of the risk 
assessment document shows that trace amounts of mercury have been 
detected in raw coke oven gas with a ratio to benzene soluble organics 
(BSO) of 2 x 10-7. Applying this ratio to the by-product 
coke plant with the highest BSO emissions (AK Steel in Ashland, 
Kentucky in Table C-5) gives an estimate of 0.002 lb/yr of mercury 
emissions from leaks. These low levels of mercury emissions show that 
mercury emissions from charging, doors, lids, and offtakes do not 
contribute significantly to the health effects posed by coke oven 
emissions from by-product coke oven batteries.
---------------------------------------------------------------------------

    \22\ Fisher, R. ``Progress in Pollution Abatement in European 
Cokemaking Industry''. Ironmaking and Steelmaking. vol. 19, no. 6., 
1992. Pages 449-456.
---------------------------------------------------------------------------

    The estimate of 27 lb/yr for the AK Steel by-product coke plant was 
not based on measurements. The company used an emission factor that was 
developed from a 1991 paper published in Germany. However, it is not in 
EPA's AP-42 compilation of emission factors, we have been unable to 
determine its basis and the type of coke battery it was developed for, 
and we cannot assess its applicability to U.S. coke batteries. We 
expect more and better data to become available in the future, and 
these data will be considered when the residual risk is assessed within 
8 years of the promulgation of the 2003 NESHAP for pushing, quenching, 
and battery stacks.
    We investigated the TRI reporting and found that most mercury 
emissions from nonrecovery batteries come from the battery stack rather 
than leaks on the battery, which are the subject of these final rule 
amendments. In addition, our examination of the TRI data reveals that 
the emissions reported by the nonrecovery coke plant (Indiana Harbor 
Coke) were overestimated and are being corrected. The plant had used an 
emission factor developed from testing an uncontrolled battery stack at 
another nonrecovery coke plant. Subsequently the company performed 
sampling of its own stack and found that its actual mercury emissions 
from the battery stack were 182 pounds per year (lb/yr). Mercury is 
emitted from the battery stack on nonrecovery batteries because there 
is no recovery of the by-products distilled from the coal; however, 
some mercury in the particulate phase is captured by the baghouse that 
is used to control emissions. These test data will be considered by EPA 
when the residual risk is evaluated for the 2003 NESHAP for pushing, 
quenching, and battery stacks.
    Finally, the commenter's assumption that mercury emissions from 
batteries are not controlled by the standard is not correct. Mercury 
emissions from leaks on the battery are controlled and regulated the 
same way as the many other volatile pollutants in raw coke oven gas. 
The ovens are inspected for leaks, and work practices are used to stop 
leaks and contain potential emissions within the gas collection system. 
Standards are in place to limit emissions from charging, doors, lids, 
and offtakes, and these standards also effectively limit emissions of 
mercury (as a volatile) and other pollutants that might otherwise occur 
if these standards were not in place.
5. Consider Other HAP
    Comment: Three commenters contended that the risk assessment is 
deficient because it did not adequately consider the risks associated 
with emissions of all HAP. One commenter stated that the 13 PBT 
constituents chosen for cancer and noncancer risk analysis inexplicably 
excluded both mercury and arsenic and that chromium and mercury were 
left out of the inhalation risk analysis. Other commenters state that 
the risk assessment must cover the carcinogenic effects of naphthalene 
and 1, 3-butadiene; coke and coal dust emissions from uncovered 
sources; and hydrogen chloride (HCl) emissions.
    Response: As stated in the risk assessment document and discussed 
in an earlier response, inhalation cancer risk from the sources covered 
by this rule was estimated using the HAP ``coke oven emissions,'' for 
which we have developed a cancer URE. It is not necessary to consider 
the presence of each constituent of the mixture of coke oven emissions 
thought to be carcinogens since their contribution to cancer risk is 
subsumed into the risk from the mixture. Section 112(f)(6) contemplates 
such an approach, as we noted in our ``Residual Risk Report to 
Congress''. In conducting the non-cancer inhalation risk assessment, we 
did use information (toxicity and emissions) for each constituent 
because there are inadequate data for a non-cancer assessment of ``coke 
oven emissions''. In general, we considered the risk due to individual 
constituents when assessing non-cancer or non-inhalation risks, when 
assessing risk

[[Page 20006]]

from emission points where the composition of the mixture may be 
different, (e.g., after the pushing emission control device), or when 
the screening level risk assessment was done. The URE for coke oven 
emissions was used for all identified process operations covered under 
the 1993 national emission standards for charging, doors, lids, and 
offtakes and for two emission sources (pushing and quenching) covered 
by the 2003 NESHAP for pushing, quenching, and battery stacks. For the 
remaining emission sources which do not emit coke oven emissions (e.g., 
the battery stack and the pushing emission control device), we selected 
constituents that had toxicity values and emissions information from 
these emission points in order to conduct an inhalation risk assessment 
or a non-inhalation, multipathway assessment. Results for the cancer 
and non-cancer risk assessment may be found in Tables A-2 through A-9 
of the risk assessment document. Multipathway results for those HAP 
selected based on our selection criteria may be found in Tables A-31 
through A-34.
    The risk assessment did not include estimates of risk for 
pollutants such as ammonia, hydrogen sulfide, coal dust, and coke dust 
because they are not listed as HAP under section 112(b). We do not read 
section 112(f) as requiring consideration of criteria pollutants and 
other pollutants which are not HAP. Section 112(f) is the corollary of 
section 112(d), which of course is directed to control of HAP. It also 
essentially adopts the pre-1990 standard for control of HAP (see, e.g., 
Legislative History page 876), which dealt exclusively with control of 
air toxics. We believe that given this linkage and prior history, 
Congress would have been explicit had it intended for us to 
dramatically change course and address risks posed by non-HAP 
pollutants under section 112(f).
    At the time the risk assessment was performed, the cancer URE for 
naphthalene was not available from the CARB, a source of toxicity 
information we use if IRIS does not have a benchmark value. Based on 
the emissions information for this HAP described in the risk assessment 
document (i.e., depending on the source, emissions of about 10 to 30 
times less than the coke oven emission estimates and a cancer URE that 
is 18 times less potent than the URE for coke oven emissions), 
naphthalene is not likely to add significantly to the cancer risk 
estimated for this source or to have an effect on the decision.
    The commenters also asked why we did not include chromium, a 
carcinogen, in the mix of carcinogens we assessed. Unlike naphthalene, 
hexavalent chromium does have a URE on IRIS, but information we 
received indicated that hexavalent chromium emissions from this process 
are unlikely due to the atomic state for this pollutant being highly 
oxidized and not conducive for forming in a chemical reducing 
atmosphere such as a coke oven. Thus, the emissions would likely be the 
trivalent chromium, which has not been shown to be carcinogenic. 
Another way to look at this issue is to assume a fixed percentage of 
total chromium is hexavalent. For example, applying the health-
protective assumption we used in our Report to Congress on Electric 
Utilities \23\ (that hexavalent chromium comprised 11 percent of the 
total chromium emissions) would result in a MIR level of approximately 
1 in a million. Therefore, it is unlikely that any chromium emissions 
from the sources considered in this source category would have any 
significant impact on the estimated total cancer risk.
---------------------------------------------------------------------------

    \23\ Table 6-1, Summary of High-End Risk Estimates from Chronic 
Inhalation Exposure of HAP for 424 U.S. Coal-Fired Utilities Based 
on the Baseline Inhalation Risk Assessment. Study of Hazardous Air 
Pollutant Emissions from Electric Utility Steam Generating Units--
Final Report to Congress. U.S. Environmental Protection Agency, 
Office of Air Quality Planning and Standards. Vol. 1. EPA 453/R-98-
004a, February 1998, page 6-3.
---------------------------------------------------------------------------

    The URE for arsenic was applied to the battery stack and the 
pushing emission control device. These emission points are the only 
ones for which we would use arsenic's specific URE in the risk 
calculations because the URE for coke oven emissions accounts for the 
cancer risk from other emission points. The highest MIR for arsenic 
from these sources was less than 1 in a million.
    Table 3-2 in the risk assessment document provides a detailed 
listing of non-cancer risks at the facility level, which includes 
estimates for arsenic and hydrogen chloride. The table shows that the 
maximum HQ for arsenic was 0.3 and was 0.00002 for hydrogen chloride. 
The non-cancer risks for chromium assuming all emissions are hexavalent 
would provide a HQ value equal to 0.01, still significantly below a 
value of 1. We believe, moreover, that this significantly overestimates 
the risk.
6. Emission Estimates
    Comment: One commenter contended that the emission estimates 
overstated HAP emissions and discussed problems with EPA's emission 
factors and calculations:
     Emissions from coke oven door leaks were overstated 
because EPA did not use the exponential model developed in the early 
1980s, overestimated the number of leaks visible from the bench and not 
the yard, and included emissions from doors with no visible leaks.
     The EPA did not adequately justify estimates of the 
frequency and severity of green pushes and understated the capture 
efficiency of pushing emission control devices. Benzene emissions from 
pushing are also overestimated.
     Emissions from battery stacks were overstated because of 
the extrapolation to higher opacities and the use of questionable test 
data for benzene.
     Emissions from by-product recovery plant process equipment 
were overstated because of the use of default values rather than a 
site-specific approach.
    Response: The issue of the exponential model developed in the early 
1980s has been discussed in great detail in the background document for 
AP-42. Relevant excerpts are summarized below:
     The theoretical model was based solely on the self-sealing 
mechanism and does not account for the current widespread use of 
supplementary sealants, new door designs, and adjusting the door seal 
to stop leaks.
     The exponential model is not applicable below 10 percent 
leaking doors, and current control levels are well below 10 percent.
     The exponential model underestimates emissions when using 
an arithmetic annual average for percent leaking doors (an exponential 
averaging of percent leaking doors must be used).
     The exponential model estimates zero emissions when no 
door leaks are visible from the yard, but we now know there are door 
leaks that cannot be seen from the yard.
     More recent sampling and analysis of door leaks of various 
sizes have provided real data on mass emission rates (as opposed to a 
theoretical and unvalidated model) and form the basis for current 
estimates.
    We used a value of 6 percent leaking doors for doors visible from 
the bench but not visible from the yard, and the commenter recommended 
a value of 3 percent based on more recent data. The value of 6 percent 
is the value recommended in AP-42 and is codified in the 1993 national 
emission standards (doors inspected from the bench under a cokeside 
shed are given a correction factor of 6 percent leaking to estimate the 
``yard'' equivalent). We acknowledge that the difference between the 
number of door leaks observed from the bench and from the yard probably 
varies from battery to battery and at the same battery

[[Page 20007]]

over time. The commenter also thought the leak rate assigned to the 
small leaks visible only from the bench was too high. However, this 
rate was based on the smallest visible leak grade (a grade of 0.5, 
which is described as a leak that is barely visible and may not be seen 
from the yard), and we cannot arbitrarily reduce it by 75 percent as 
the commenter suggested. We included the variability associated with 
leaks visible only from the bench and the variability in leak rates in 
our uncertainty analysis. We think that analysis places reasonable 
bounds on our emission estimates.
    We did not include emissions from leaks that were not visible in 
the emission estimates used in the risk assessment. The potential for 
emissions from leaks that are not visible was factored into the 
uncertainty analysis and indicated that actual emissions could be 
higher than we estimated. However, we also acknowledged that emissions 
could be lower than we estimated.
    The frequency and severity of green pushes used in the risk 
analysis (not part of the source category at issue) are explained in 
detail in the background information document for pushing, quenching, 
and battery stacks (Docket Item OAR-2003-0051-0085). The document 
estimates the frequency of green pushes once the 2003 NESHAP for 
pushing, quenching, and battery stacks is fully implemented. 
Admittedly, better estimates can be made in the future based on actual 
performance data generated after the compliance date of the final rule 
amendments. The projections of methylene chloride soluble organics 
(MCSO) emissions are based on the performance of the best-performing 
batteries that were used to develop the MACT floor. Data for 3,700 
observations from 15 batteries that were the best performers had only 
one severely green push with an opacity exceeding 50 percent. Two other 
batteries that will have to improve their performance to meet the 
standard had 2 percent green pushes. A best estimate of 0.5 percent 
severely green pushes was judged likely to be an overestimate once all 
batteries were subject to the standard. For moderately green pushes in 
the range of 30 to 50 percent opacity, the best-controlled batteries 
averaged 0 percent to 5 percent of the pushes in this range (13 out of 
3,700 observations). An upper-bound estimate of 5 percent was used for 
moderately green pushes.
    A capture efficiency of 10 percent for a severely green push is 
based on observations that most of the emissions escape capture during 
pushing and the fact that heavy emissions (some observed at 90 to 100 
percent opacity) continue during travel to the quench tower when there 
is no hood to capture any of the emissions. During a push that is not 
green, some emissions escape capture and again none of the emissions 
during travel are captured; consequently, an estimate of 90 percent 
capture seems reasonable for that case. A best estimate of 40 percent 
capture was used for moderately green pushes.
    The benzene emission factor used for pushing is 2.4 x 
10-\4\ lb/ton of coke based on three runs at one plant 
producing blast furnace coke. The commenter submitted data from a plant 
producing foundry coke that showed benzene emissions were less than 9 x 
10-\5\ lb/ton, a factor of about two lower. The amount of 
benzene emitted from pushing will depend on how green the coke is, and 
a push that is fully coked would have very little benzene. It is 
difficult to determine which test is most representative, and the 
benzene emissions can be expected to be quite variable from push to 
push. However, even with the higher emission factor, benzene emissions 
from pushing were not significant in the risk analysis (e.g., less than 
100 lb/yr).
    The commenter stated that the extrapolation of test results for 
battery stacks based on opacity is unsupportable because there is no 
established relationship between opacity and HAP. As explained in the 
background document, the battery stack that was tested had a very low 
opacity (1.7 percent), but the 2003 NESHAP are expected to achieve an 
average opacity of 5 percent for battery stacks. Consequently, applying 
the test results for this one battery to all other batteries to 
estimate the emissions once the 2003 NESHAP become effective could 
underestimate emissions. Although no correlation has been firmly 
established between opacity and HAP, there is an established 
relationship between opacity and mass concentration of particles. In 
addition, sampling and analysis has shown that the PM in battery stack 
emissions contains HAP, including organic PM and PAH. (These PAH are a 
primary constituent of coke oven emissions, the primary HAP evaluated 
in the risk assessment.) Consequently, battery stack emissions were 
scaled from 1.7 to 5 percent opacity to avoid underestimating emissions 
from other batteries once the standard is implemented.
    The commenter stated that EPA used the results from the two highest 
of four tests to estimate benzene emissions from battery stacks and 
that using the average of all four tests would have resulted in 
emissions that were 40 percent lower. The results for benzene in parts 
per million (ppm) for the four tests were 0.1 to 0.2, 0.6 to 1.6, 1.8 
to 4.1, and 2.6 to 3.2. One of the four tests is an order of magnitude 
less than the others and appears to be an outlier. The average values 
of the other three tests are 1, 3, and 3 ppm. We used a value of 3 ppm 
because it is the statistical mode (most frequently occurring test 
average), it is representative of two of the four tests, and this value 
would not tend to underestimate emissions. Using the average value for 
all four tests would have resulted in an emissions estimate 40 percent 
lower than our original estimate. However, even if our original 
estimate overestimates emissions, there were no significant adverse 
health effects estimated for this source for benzene. In addition, EPA 
will re-evaluate the emissions and risks from battery stacks within 8 
years after the promulgation date of the MACT standard for pushing, 
quenching, and battery stacks. At that time, the emission estimate will 
be revised based on additional test data that become available.
    Benzene emissions from process equipment in the by-product recovery 
plant were estimated from AP-42 emission factors, site-specific 
information on the processes, and their capacities. The commenter 
recommended using EPA's TANKS model with detailed site-specific 
information to estimate emissions because it would be more accurate and 
emissions would be lower. However, the AP-42 emission factors that we 
used have been widely accepted and used in other contexts, and they 
account for sources that have controls in place. We did not have 
detailed and verifiable information for the numerous site-specific 
factors that would be needed to use the TANKS model. We agree with the 
commenter that the use of TANKS is an acceptable alternative when such 
details are available and the model is applicable to the emission point 
of interest. However, there are some process vessels in the by-product 
plant where the model is not applicable because it does not fully 
account for the emission mechanism, such as tanks that are heated or 
purged and have a vapor flow other than from working and breathing 
losses, uncovered tanks, those for which there is no good estimate of 
the vapor phase concentration, and condensers.
    The commenter pointed out that we used site-specific monitoring 
data to estimate benzene emissions from equipment leaks for all plants 
except one (Tonawanda Coke) and that the emission factors applied to 
this plant overestimated emissions. We requested

[[Page 20008]]

site-specific monitoring data from all plants to estimate emissions, 
but we did not receive such information from Tonawanda Coke. We agree 
that generally the site-specific approach provides emission estimates 
lower than those from the default emission factors. Our emission 
estimates were health protective, and even with a tendency to 
overestimate benzene emissions from Tonawanda Coke, the estimated risk 
from these benzene emissions is low.

C. Comments on Section 112(d)(6) Review Policy

1. Approach for Existing Sources
    Comment: Eight commenters agreed that a new analysis of MACT floors 
for existing sources is not part of the 8-year review requirement. As 
EPA concluded, such periodic re-determination of the MACT floor would 
effectively convert existing source requirements into new source 
requirements. In support, one commenter pointed to the plain language 
of CAA section 112(d)(6), the legislative history, similar review 
requirements under sections 109 and 111, and the absence of 
Congressional intent for new floor analyses.
    Two commenters disagreed with EPA's conclusions. One commenter 
explained that the MACT floor provisions in section 112(d)(3) give 
meaning to the phrase ``emission standards promulgated under this 
section'' in section 112(d)(6) so that EPA is obligated to do a new 
floor analysis when revising the standards for existing sources. In 
addition, EPA's argument (that omission of the term ``emission 
limitation achieved'' suggests that no additional floor determination 
is required) ignores the statutory text. There is no need to include 
the floor language in section 112(d)(6) since section 112(d)(3) already 
ensured that any existing source standard would meet the floor 
requirements. The EPA's other argument (that additional floor analyses 
would effectively convert existing source standards into new source 
standards) is unreasonable and not necessarily true because EPA could 
find that sources do not perform better than the floor level of 
control. If facilities developed methods to reduce HAP emissions in the 
previous 8 years, requiring all sources in the category to achieve 
similar control would be consistent with Congressional intent under 
section 112 and the specific direction given in section 112(d)(6).
    Response: Section 112(d)(6) requires us to ``* * * review, and 
revise as necessary (taking into account developments in practices, 
processes, and control technologies), emission standards promulgated 
under this section.'' The provision does not mandate that this review 
be conducted in a single, unvarying manner, other than having to take 
into account ``developments in practices, processes, and control 
technologies.''
    The commenter maintained that because of the reference to 
``emission standards promulgated under this section,'' we are 
necessarily required to repeat the section 112(d) standard development 
process which includes re-determining MACT floors. A more natural 
reading of the provision is that we are to review the section 112(d) 
standards considering developments in practices, processes, and control 
technologies. EPA may then, in its discretion, amend the standards if 
the agency concludes such action is necessary. Indeed, we believe that 
this is the meaning Congress intended, since section 112(d)(6) 
originated in House and Senate Committee provisions that predated 
introduction of the MACT floor language, and mirrors routine periodic 
reevaluation requirements found in other statutory provisions requiring 
technology-based standards. Moreover, we reiterate that there is no 
indication that Congress intended for section 112(d)(6) to inexorably 
force existing source standards progressively lower and lower in each 
successive review cycle, the likely result of requiring successive 
floor determinations (69 FR 48351).
    We note that with respect to revision of standards for new sources, 
the section 112(d)(6) analysis of practices, processes, and control 
technologies, and costs and emission reductions associated with those 
technologies (conducted as part of the determination of whether 
different standards are necessary), may indicate that revised standards 
for new sources are warranted. The final rule amendments do not adopt 
different standards for new by-product batteries. New by-product 
batteries would be required to meet zero leak standards for doors, lid, 
and offtakes unless a new by-product technology (such as operation of 
the ovens under negative pressure) is developed. The by-product battery 
technology currently in use cannot achieve zero leaks; consequently, 
new coke batteries would likely be nonrecovery batteries, which have 
been the only type of new battery constructed in the past 20 years. We 
are amending the charging limit for new nonrecovery batteries to 
reflect new technical developments (69 FR 48351). These changes can be 
readily incorporated at new sources with minimal cost.
2. Relationship Between Residual Risk Standards and Review Requirements
    Comment: Six commenters stated that once EPA promulgates a standard 
that provides an ample margin of safety, the review requirement under 
section 112(d)(6) is satisfied. One commenter stated that Congress 
intended the section 112(d)(6) review to update the underlying 
technology-based standards irrespective of residual risk.
    Response: We begin by noting ambiguity in the text and placement of 
section 112(d)(6). The obligation to periodically review (and possibly 
update) emissions standards applies to standards promulgated under 
``this section.'' A possible reading of the word ``section'' is that 
the periodic review obligation applies not only to emissions standards 
adopted under section 112(d), but also to emissions standards adopted 
under any other provision of section 112, including section 112(f) 
(note that section 112(f)(2) is entitled ``emissions standards''). On 
the other hand, section 112(d)(6) is placed in the context of section 
112(d) generally, which deals only with technology based ``MACT'' 
standards. This placement could be construed as requiring the periodic 
review obligation to only apply to emissions standards adopted under 
section 112(d).
    We resolve this ambiguity by concluding that section 112(d)(6) 
should be interpreted as applying only to standards adopted under 
section 112(d). This conclusion is based on several factors. First, all 
of the other provisions of section 112(d) are specific to the 
obligation to adopt technology standards. It would be inconsistent with 
the structure of section 112(d) as a whole to conclude that section 
112(d)(6) should be construed to apply more broadly than all of the 
other companion provisions in section 112(d).
    Second, it is natural to assume that the technology on which a 
particular section 112(d) standard is based could evolve over time and 
allow EPA, as appropriate, to update the standard to reflect the 
evolving technology. Other text in section 112(d)(6) is clearly focused 
on this possibility of technological innovation (``* * * taking into 
account developments in practices, processes, and control technologies 
* * *''). In contrast, the basic obligation under section 112(f) is to 
make sure that public health risks due to emissions from a category or 
subcategory provide an ample margin of safety. Technology (and the 
possibility that technology will improve over time) remains relevant 
under section 112(f), but only for the purpose of determining an 
appropriate

[[Page 20009]]

ample margin of safety. Notably, technology is only one of many factors 
that may be relevant in determining the ample margin of safety. Thus, 
evolving technology--which is the clear focus of section 112(d)(6)--is 
central to the purposes of section 112(d), while it is only one 
consideration among many that may be relevant under section 112(f). If 
Congress had intended section 112(d)(6) to encompass section 112(f), a 
broader range of considerations would logically have been mandated for 
the periodic review.
    Finally, we believe our interpretation is supported by legislative 
history. The genesis of section 112(d)(6) can be traced to earlier 
bills passed by the Senate and the House, all of which made it clear 
that the periodic review applied to section 112(d) MACT standards.\24\ 
Of particular weight is the Report of the Senate Committee on 
Environment and Public Works on the Clean Air Act Amendments of 1989 
that clarifies that the section 112(d)(6) review provisions were 
intended to apply to MACT standards: ``The Administrator is to review 
and revise emission standards promulgated under section 112(d) no less 
than every seven years.''
---------------------------------------------------------------------------

    \24\ See S. 1894, Clean Air Standards Attainment Act of 1987; S. 
1630, Clean Air Act Amendments of 1989; and H.R. 3030, Clean Air Act 
Amendments of 1990.
---------------------------------------------------------------------------

    Having said that, we believe that the findings that underlie a 
section 112(f) determination should be key factors in making any 
subsequent section 112(d)(6) determinations for the related section 
112(d) standard. For example, if the ample margin of safety analysis 
for the section 112(f) standard was not based at all on the 
availability or cost of particular control technologies, then advances 
in air pollution control technology should not justify revising the 
MACT standard pursuant to section 112(d)(6) because the section 112(f) 
standard would continue to assure an adequate level of safety. 
Similarly, if the ample margin of safety analysis for a section 112(f) 
standard shows that remaining risk for non-threshold pollutants falls 
below 1 in a million and for threshold pollutants falls below a similar 
threshold of safety, then no further revision would be needed because 
an ample margin of safety has already been assured.

D. Specific Comments on Section 112(d)(6) Review of Coke Ovens

1. Nonrecovery Technology
    Comment: One commenter stated that EPA admitted that risk levels 
could be reduced substantially with nonrecovery technology. However, 
EPA decided not to require this technology because the costs of 
replacing existing batteries with nonrecovery batteries would be 
financially crippling to the industry. Although EPA provided some cost 
estimates, the Agency did not say why that cost would be crippling to 
the industry or even to the individual companies involved. Rather, EPA 
explained that the industry is currently depressed and plants might 
choose to shut down. The EPA must substantiate its claims.
    Response: We explained at proposal that replacing existing 
batteries with nonrecovery batteries would be financially crippling 
because the construction of a nonrecovery battery requires a capital 
investment on the order of hundreds of millions of dollars (about $300 
per ton of coke capacity). For example, the estimated capital cost to 
replace batteries on the MACT track ranges from $50 to $290 million per 
plant based on the existing coke capacity at these plants. Based on 
recent trends that show a continuing decline in domestic coke capacity 
due to shutdowns, these coke facilities would be more likely to 
permanently close rather than construct new nonrecovery batteries. For 
example, 12 of the 30 coke plants operating in 1993 have permanently 
shut down, and five of these plants were on the MACT track. 
Consequently, we determined that requiring the replacement of existing 
batteries with nonrecovery batteries was not a reasonable or 
economically feasible option.
    We also examined the ability of the companies involved to recoup 
their investment if they were to replace existing batteries with 
nonrecovery batteries. The four existing by-product coke plants on the 
MACT track are owned by two companies: AK Steel, which produces furnace 
coke for internal consumption, and Tonawanda Coke Corporation, which 
produces 15 to 20 percent of the foundry coke sold in the U.S. Based on 
the Quarterly Financial Report from the U.S. Bureau of the Census,\25\ 
the average return on sales for all reporting companies within the iron 
and steel industries from 2nd Quarter 2003 to 2nd Quarter 2004 ranged 
from negative 5.9 percent to 9.8 percent. The weighted average price of 
coke is approximately $120 per short ton. Using the highest profit rate 
in 2004 (which is optimistic), the implied profit per short ton is 
approximately $12 per short ton. Our conclusion is that with a 7 
percent discount rate, companies would not able to recoup investment 
for a nonrecovery battery (approximately $300 per ton). Even a 50-year 
time profile at this profit level would not be sufficient to offset the 
investment. Therefore these coke facilities would be more likely to 
permanently close rather than construct new nonrecovery batteries. 
These closures could have industry wide implications, particularly for 
the foundry coke market, since Tonawanda accounts for a significant 
share of foundry coke production in the U.S.
---------------------------------------------------------------------------

    \25\ Table 4, Quarterly Financial Report for Manufacturing, 
Mining, and Trade Corporations. U.S. Bureau of the Census, Second 
Quarter, Series QFR 04-2Q. 2004.
---------------------------------------------------------------------------

2. Lack of New Requirements
    Comment: One commenter believed that the proposed amendments were 
deficient because they contained no new requirements despite the 
remaining risk from facilities.
    Response: The commenter is incorrect--the final rule amendments are 
new and provide more stringent requirements for the MACT track 
batteries. The limit for leaking doors decreases from 5 percent to 4 
percent for foundry coke batteries and to 3.3 percent for other 
batteries, the limit for lid leaks decreases from 0.6 percent to 0.4 
percent, and the limit for offtake leaks decreases from 3 percent to 
2.5 percent. The standard for new batteries and for reconstructed 
batteries if there is an increase in capacity is already quite 
stringent. Except for batteries utilizing a new by-product recovery 
technology (such as by-product ovens operated under negative pressure), 
the standard is 0 percent leaking doors, lids, and offtakes. The 
current by-product battery technology cannot achieve this level of 
control; consequently, new batteries are likely to use the nonrecovery 
technology. In fact, the only new batteries constructed over the past 
20 years have been nonrecovery batteries.
3. Charging Limit for Nonrecovery Batteries
    Comment: One commenter requested that the proposed limit for 
charging (20 percent opacity for five consecutive charges) in 40 CFR 
63.303(d) also apply to existing nonrecovery batteries, not just new 
batteries as proposed. As proposed, the charging limit would not apply 
to nonrecovery batteries in the commenter's state (including one 
existing plant and a new plant for which construction began before the 
date of proposal).
    Response: We based our proposal for more stringent standards for 
new sources on the performance of the best-controlled source, and this 
plant was developing an improved capture system

[[Page 20010]]

for charging emissions. We concluded that it was not appropriate to 
increase the stringency of the current NESHAP for already-operating 
nonrecovery batteries. This limit is appropriate for new sources, which 
are those constructed after the date of proposal of these final rule 
amendments, because it allows the new requirements to be incorporated 
into the considerations of design and operation of the new source. 
Further, we believe that the quantified limits on PM which two of the 
already-operating nonrecovery batteries are achieving (69 FR 48351-
48352) can be readily (and appropriately) incorporated in these 
batteries' operating permits as part of the State implementation plan 
process. The suggestion by the commenter that we use this rulemaking to 
amend the standard for these batteries to lock in their level of 
performance thus appears to be unnecessary.
4. Costs
    Comment: Two commenters asked EPA to avoid characterizing the costs 
of $4,500/yr as ``small,'' ``minimal'' and ``very little.'' The 
additional reduction that would be achieved is the last increment in a 
series of reductions made by a distressed industry. The commenters 
stated that, in their opinion, the incremental cost effectiveness is 
actually high ($45,000 per ton), and the costs should be presented in 
this format. They stated that the EPA should also recognize the 
industry's success and overall cost in reducing emissions to meet the 
stringent level of control.
    Response: The original 1993 national emission standards resulted in 
oven repairs, increased maintenance, and better work practices that 
have reduced emissions to allow batteries to meet a more stringent 
level of control. All of these activities have resulted in increased 
costs for the control of emissions, although the emission reduction 
benefits are substantial. In addition, the 1993 national emission 
standards require daily monitoring to identify leaks, and the data show 
the industry's success in reducing emissions.
    We believe the cost of complying is reasonable considering that an 
estimated 200,000 fewer people will be exposed to risks greater than 1 
in a million, and the annual cancer incidence would be reduced by 0.03. 
We agree with the commenters that the estimate of $4,500/yr is the most 
recent increment in a series of reductions, but remain steadfast in our 
belief that this number is minimal.

IV. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review

    Under Executive Order 12866 (58 FR 51735, October 4, 1993), the EPA 
must determine whether the regulatory action is ``significant'' and, 
therefore, subject to review by the Office of Management and Budget 
(OMB) and the requirements of the Executive Order. The Executive Order 
defines a ``significant regulatory action'' as one that is likely to 
result in a rule that may:
    (1) Have an annual effect on the economy of $100 million or more or 
adversely affect in a material way the economy, a sector of the 
economy, productivity, competition, jobs, the environment, public 
health or safety, or State, local, or tribal governments or 
communities;
    (2) Create a serious inconsistency or otherwise interfere with an 
action taken or planned by another agency;
    (3) Materially alter the budgetary impact of entitlement, grants, 
user fees, or loan programs or the rights and obligations of recipients 
thereof; or
    (4) Raise novel legal or policy issues arising out of legal 
mandates, the President's priorities, or the principles set forth in 
the Executive Order.
    Under the terms of Executive Order 12866, it has been determined 
that this regulatory action is a ``significant regulatory action'' 
because it raises novel legal or policy issues. As such, this action 
was submitted to OMB for Executive Order 12866 review. Changes made in 
response to OMB suggestions or recommendations will be documented in 
the public record.

B. Paperwork Reduction Act

    The information collection requirements in the final rule 
amendments have been submitted for approval to OMB under the Paperwork 
Reduction Act, 44 U.S.C. 3501 et seq. The information collection 
request (ICR) prepared by EPA has been assigned EPA ICR No. 1362.07. 
The information collection requirements are not enforceable until OMB 
approves them.
    The information requirements are based on notification, 
recordkeeping, and reporting requirements in the NESHAP General 
Provisions (40 CFR part 63, subpart A), which are mandatory for all 
operators subject to national emission standards. These recordkeeping 
and reporting requirements are specifically authorized by section 114 
of the CAA (42 U.S.C. 7414). All information submitted to EPA pursuant 
to the recordkeeping and reporting requirements for which a claim of 
confidentiality is made is safeguarded according to Agency policies set 
forth in 40 CFR part 2, subpart B.
    The final rule amendments establish work practice requirements 
designed to improve control of door leaks applicable to all nonrecovery 
coke oven batteries. The owner or operator also is required to add 
certain information on malfunctions associated with door leaks to the 
startup, shutdown, and malfunction plan. New nonrecovery batteries also 
are required to implement the same work practice standards that already 
apply to existing nonrecovery batteries. Plant owners or operators are 
required to submit an initial notification of compliance status and 
semiannual compliance reports. Records are required to demonstrate 
compliance with applicable emission limitations and work practice 
requirements. Additional requirements apply to a new nonrecovery coke 
oven battery, but none are expected during the 3-year period of this 
ICR. This action does not impose any new or revised information 
collection burden on by-product coke oven batteries subject to the 
final rule amendments. These batteries are currently meeting the 
monitoring, recordkeeping, and reporting requirements in the 1993 
national emission standards.
    The increased annual average monitoring, reporting, and 
recordkeeping burden for this collection (averaged over the first 3 
years of the ICR) is estimated to total 448 labor hours per year at a 
cost of $28,338. This includes an increase of three responses per year 
from one respondent for an average of about 148 hours per response. No 
capital/startup costs or operation and maintenance costs are associated 
with the monitoring requirements.
    Burden means the total time, effort, or financial resources 
expended by persons to generate, maintain, retain, or disclose or 
provide information to or for a Federal agency. This includes the time 
needed to review instructions; develop, acquire, install, and utilize 
technology and systems for the purposes of collecting, validating, and 
verifying information, processing and maintaining information, and 
disclosing and providing information; adjust the existing ways to 
comply with any previously applicable instructions and requirements; 
train personnel to be able to respond to a collection of information; 
search data sources; complete and review the collection of information; 
and transmit or otherwise disclose the information.
    An agency may not conduct or sponsor, and a person is not required 
to

[[Page 20011]]

respond to, a collection of information unless it displays a currently 
valid OMB control number. The OMB control numbers for EPA's regulations 
in 40 CFR part 63 are listed in 40 CFR part 9.

C. Regulatory Flexibility Act

    The EPA has determined that it is not necessary to prepare a 
regulatory flexibility analysis in connection with the final rule 
amendments. For the purposes of assessing the impacts of today's final 
rule amendments on small entities, small entity is defined as: (1) A 
small business as defined by the Small Business Administrations' 
regulations at 13 CFR 121.201; (2) a small governmental jurisdiction 
that is a government of a city, county, town, school district or 
special district with a population of less than 50,000; and (3) a small 
organization that is any not-for-profit enterprise which is 
independently owned and operated and that is not dominant in its field.
    After considering the economic impacts of today's final rule 
amendments on small entities, EPA has concluded that this action will 
not have a significant economic impact on a substantial number of small 
entities. We have determined that of the five companies subject to the 
requirements of the final rule amendments, one company (operating a 
total of three batteries) is considered a small entity but it will 
experience no significant additional regulatory costs because it is 
already meeting the stricter emissions limitations for by-product coke 
oven batteries included in the final rule amendments, as well as the 
monitoring, recordkeeping, and reporting requirements.
    Although the final rule amendments will not have a significant 
economic impact on a substantial number of small entities, we 
nonetheless tried to reduce the impact of the final rule amendments on 
small entities. Prior to proposal, we held meetings with industry trade 
associations and company representatives to discuss the amendments and 
have included provisions that address their concerns.

D. Unfunded Mandates Reform Act

    Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), Public 
Law 104-4, establishes requirements for Federal agencies to assess the 
effects of their regulatory actions on State, local, and tribal 
governments and the private sector. Under section 202 of the UMRA, the 
EPA generally must prepare a written statement, including a cost-
benefit analysis, for proposed and final rules with ``Federal 
mandates'' that may result in expenditures by State, local, and tribal 
governments, in the aggregate, or by the private sector, of $100 
million or more in any 1 year. Before promulgating an EPA rule for 
which a written statement is needed, section 205 of the UMRA generally 
requires the EPA to identify and consider a reasonable number of 
regulatory alternatives and adopt the least costly, most cost-
effective, or least-burdensome alternative that achieves the objectives 
of the rule. The provisions of section 205 do not apply when they are 
inconsistent with applicable law. Moreover, section 205 allows the EPA 
to adopt an alternative other than the least-costly, most cost-
effective, or least-burdensome alternative if the Administrator 
publishes with the final rule an explanation why that alternative was 
not adopted. Before the EPA establishes any regulatory requirements 
that may significantly or uniquely affect small governments, including 
tribal governments, it must have developed under section 203 of the 
UMRA a small government agency plan. The plan must provide for 
notifying potentially affected small governments, enabling officials of 
affected small governments to have meaningful and timely input in the 
development of EPA regulatory proposals with significant Federal 
intergovernmental mandates, and informing, educating, and advising 
small governments on compliance with the regulatory requirements.
    EPA has determined that the final rule amendments do not contain a 
Federal mandate that may result in expenditures of $100 million or more 
for State, local, and tribal governments, in the aggregate, or to the 
private sector in any 1 year. No significant costs are attributable to 
the final rule amendments. Thus, the final rule amendments are not 
subject to the requirements of sections 202 and 205 of the UMRA. In 
addition, the final rule amendments do not significantly or uniquely 
affect small governments because they contain no requirements that 
apply to such governments or impose obligations upon them. Therefore, 
the final rule amendments are not subject to section 203 of the UMRA.

E. Executive Order 13132: Federalism

    Executive Order 13132 (64 FR 43255, August 10, 1999) requires EPA 
to develop an accountable process to ensure ``meaningful and timely 
input by State and local officials in the development of regulatory 
policies that have federalism implications.'' ``Policies that have 
federalism implications'' is defined in the Executive Order to include 
regulations that have ``substantial direct effects on the States, on 
the relationship between the national government and the States, or on 
the distribution of power and responsibilities among the various levels 
of government.''
    The final rule amendments do not have federalism implications. They 
will not have substantial direct effects on the States, on the 
relationship between the national government and the States, or on the 
distribution of power and responsibilities among the various levels of 
government, as specified in Executive Order 13132. None of the affected 
plants are owned or operated by State governments. Thus, Executive 
Order 13132 does not apply to the final rule amendments.

F. Executive Order 13175: Consultation and Coordination With Indian 
Tribal Governments

    Executive Order 13175 (65 FR 67249, November 6, 2000) requires EPA 
to develop an accountable process to ensure ``meaningful and timely 
input by tribal officials in the development of regulatory policies 
that have tribal implications.'' ``Policies that have tribal 
implications'' is defined in the Executive Order to include regulations 
that have ``substantial direct effects on one or more Indian tribes, on 
the relationship between the Federal government and Indian tribes.''
    The final rule amendments do not have tribal implications, as 
specified in Executive Order 13175. They will not have substantial 
direct effects on tribal governments, on the relationship between the 
Federal government and Indian tribes, or on the distribution of power 
and responsibilities between the Federal government and Indian tribes. 
No tribal governments own plants subject to the MACT standards for coke 
oven batteries. Thus, Executive Order 13175 does not apply to the final 
rule amendments.

G. Executive Order 13045: Protection of Children From Environmental 
Health & Safety Risks

    Executive Order 13045 (62 FR 19885, April 23, 1997) applies to any 
rule that: (1) Is determined to be ``economically significant,'' as 
defined under Executive Order 12866, and (2) concerns an environmental 
health or safety risk that EPA has reason to believe may have a 
disproportionate effect on children. If the regulatory action meets 
both criteria, the EPA must evaluate the environmental health or safety 
effects of the planned rule on children and explain why the planned 
regulation is preferable to other potentially effective and reasonably 
feasible alternatives considered by the Agency.

[[Page 20012]]

    While these final rule amendments are not subject to the Executive 
Order because they are not economically significant as defined in 
Executive Order 12866, this rule is relevant under Executive Order 
13045 because it represents the first application of the Agency's 
``Supplemental Guidance for Assessing Susceptibility from Early-Life 
Exposure to Carcinogens.'' In particular, the Supplemental Guidance 
addresses the potential of an increased susceptibility to developing 
cancers that may occur later in life associated with exposure to 
compounds with a mutagenic mode of action in the early-life years. 
Following the Agency's Supplemental Guidance for compounds that act 
through a mutagenic mode of action, we have applied a default 
adjustment factor in developing estimates of lifetime cancer risks in 
this rulemaking to account for any potential susceptibility that may be 
due to early-life or childhood exposure. The results of this assessment 
are contained in section I of this preamble.

H. Executive Order 13211: Actions That Significantly Affect Energy 
Supply, Distribution, or Use

    The final rule amendments are not a ``significant energy action'' 
as defined in Executive Order 13211 (66 FR 28355, May 22, 2001) because 
they are not likely to have a significant adverse effect on the supply, 
distribution, or use of energy. Further, we have concluded that the 
final rule amendments are not likely to have any adverse energy 
impacts.

I. National Technology Transfer Advancement Act

    Section 112(d) of the National Technology Transfer and Advancement 
Act (NTTAA) of 1995 (Pub. L. No. 104-113; 15 U.S.C. 272 note) directs 
the EPA to use voluntary consensus standards (VCS) in their regulatory 
and procurement activities unless to do so would be inconsistent with 
applicable law or otherwise impracticable. VCS are technical standards 
(e.g., material specifications, test methods, sampling procedures, 
business practices) developed or adopted by one or more voluntary 
consensus bodies. The NTTAA requires EPA to provide Congress, through 
the OMB, explanations when the Agency decides not to use available and 
applicable VCS.
    The final rule amendments involve technical standards. The final 
rule amendments use EPA Methods 1, 2, 2F, 2G, 3, 3A, 3B, 4, 5, 5D (PM) 
and 9 (opacity) of 40 CFR part 60, appendix A.
    Consistent with the NTTAA, we conducted searches to identify VCS in 
addition to these EPA methods. No applicable VCS were identified for 
EPA Methods 2F, 2G, 5D, and 9. One VCS was identified as an acceptable 
alternative to EPA test methods for the purposes of the final rule 
amendments. The ASME PTC 19-10-1981--Part 10, ``Flue and Exhaust Gas 
Analyses,'' (incorporated by reference) is cited in the final rule 
amendments for its manual method for measuring the oxygen, carbon 
dioxide, and carbon monoxide content of exhaust gas. This part of ASME 
PTC 19-10-1981--Part 10 is an acceptable alternative to Method 3B.
    Our search for emissions monitoring procedures identified fourteen 
VCS applicable to the final rule amendments. The EPA determined that 
twelve of the VCS identified for measuring PM were impractical 
alternatives to EPA test methods due to lack of equivalency, detail, 
specific equipment requirements, or quality assurance/quality control 
requirements. The two remaining VCS identified in the search were not 
available at the time the review was conducted because they are under 
development by a voluntary consensus body: ASME/BSR MFC 13M, ``Flow 
Measurement by Velocity Traverse,'' for EPA Method 2 (and possibly 
Method 1) and ASME/BSR MFC 12M, ``Flow in Closed Conduits Using 
Multiport Averaging Pitot Primary Flowmeters,'' for EPA Method 2. 
Therefore, EPA did not adopt those VCS for this purpose. Detailed 
information on the EPA's search and review results is included in the 
docket.
    Sections 63.309(j) through (l) of the final rule amendments list 
the EPA test methods that are required. Under 40 CFR 63.7(f) and 40 CFR 
63.8(f), a source may apply to EPA for permission to use alternative 
test methods or monitoring requirements in place of any of the EPA test 
methods, performance specifications, or procedures.

J. Congressional Review Act

    The Congressional Review Act, 5 U.S.C. 801 et seq., as added by the 
Small Business Regulatory Enforcement Act of 1996, generally provides 
that before a rule may take effect, the agency promulgating the rule 
must submit a rule report, which includes a copy of the rule, to each 
House of the Congress and to the Comptroller General of the United 
States. The EPA will submit a report containing the final rule 
amendments and other required information to the U.S. Senate, the U.S. 
House of Representatives, and the Comptroller General of the United 
States prior to publication of the final rule amendments in the Federal 
Register. A major rule cannot take effect until 60 days after it is 
published in the Federal Register. This action is not a ``major rule'' 
as defined by 5 U.S.C. 804(2). The final rule amendments will be 
effective on April 15, 2005.

List of Subjects in 40 CFR Part 63

    Environmental protection, Air pollution control, Hazardous 
substances, Incorporation by reference, Reporting and recordkeeping 
requirements.

    Dated: March 31, 2005.
Stephen L. Johnson,
Acting Administrator.


0
For the reasons stated in the preamble, title 40, chapter I, part 63 of 
the Code of Federal Regulations is amended as follows:

PART 63--[AMENDED]

0
1. The authority citation for part 63 continues to read as follows:

    Authority: 42 U.S.C. 7401 et seq.

Subpart A--[Amended]

0
2. Section 63.14 is amended by revising paragraph (i)(3) to read as 
follows:


Sec.  63.14  Incorporations by reference.

* * * * *
    (i) * * *
    (3) ANSI/ASME PTC 19.10-1981, ``Flue and Exhaust Gas Analyses [Part 
10, Instruments and Apparatus],'' IBR approved for Sec. Sec.  
63.309(k)(1)(iii), 63.865(b), 63.3166(a)(3), 63.3360(e)(1)(iii), 
63.3545(a)(3), 63.3555(a)(3), 63.4166(a)(3), 63.4362(a)(3), 
63.4766(a)(3), 63.4965(a)(3), 63.5160(d)(1)(iii), 63.9307(c)(2), and 
63.9323(a)(3) and Table 5 to Subpart DDDDD of this part.
* * * * *

Subpart L--[Amended]

0
3. Section 63.300 is amended as follows:
0
a. Redesignating existing paragraphs (a)(3) through (a)(5) as (a)(5) 
through (a)(7); and
0
b. Adding new paragraphs (a)(3), and (a)(4).


Sec.  63.300  Applicability.

    (a) * * *
    (3) July 14, 2005, for existing by-product coke oven batteries 
subject to emission limitations in Sec.  63.302(a)(3) and for 
nonrecovery coke oven batteries subject to the emission limitations and 
requirements in Sec.  63.303(b)(3) or (c);
    (4) Upon startup for a new nonrecovery coke oven battery subject to 
the emission limitations and

[[Page 20013]]

requirements in Sec.  63.303(b), (c), and (d). A new nonrecovery coke 
oven battery subject to the requirements in Sec.  63.303(d) is one for 
which construction or reconstruction commenced on or after August 9, 
2004;
* * * * *

0
4. Section 63.302 is amended by adding new paragraph (a)(3) to read as 
follows:


Sec.  63.302  Standards for by-product coke oven batteries.

    (a) * * *
    (3) On and after July 14, 2005;
    (i) 4.0 percent leaking coke oven doors for each tall by-product 
coke oven battery and for each by-product coke oven battery owned or 
operated by a foundry coke producer, as determined by the procedures in 
Sec.  63.309(d)(1);
    (ii) 3.3 percent leaking coke oven doors for each by-product coke 
oven battery not subject to the emission limitation in paragraph 
(a)(3)(i) of this section, as determined by the procedures in Sec.  
63.309(d)(1);
    (iii) 0.4 percent leaking topside port lids, as determined by the 
procedures in Sec.  63.309(d)(1);
    (iv) 2.5 percent leaking offtake system(s), as determined by the 
procedures in Sec.  63.309(d)(1); and
    (v) 12 seconds of visible emissions per charge, as determined by 
the procedures in Sec.  63.309(d)(2).
* * * * *

0
5. Section 63.303 is amended as follows:
0
a. Redesignating paragraphs (b)(3) and (b)(4) as (b)(4) and (b)(5);
0
b. Adding new paragraph (b)(3); and
0
c. Adding new paragraphs (c) and (d).


Sec.  63.303  Standards for nonrecovery coke oven batteries.

* * * * *
    (b) * * *
    (3) For charging operations, the owner or operator shall implement, 
for each day of operation, the work practices specified in Sec.  
63.306(b)(6) and record the performance of the work practices as 
required in Sec.  63.306(b)(7).
* * * * *
    (c) Except as provided in Sec.  63.304, the owner or operator of 
any nonrecovery coke oven battery shall meet the work practice 
standards in paragraphs (c)(1) and (2) of this section.
    (1) The owner or operator shall observe each coke oven door after 
charging and record the oven number of any door from which visible 
emissions occur. Emissions from coal spilled during charging or from 
material trapped within the seal area of the door are not considered to 
be a door leak if the owner or operator demonstrates that the oven is 
under negative pressure, and that no emissions are visible from the top 
of the door or from dampers on the door.
    (2) Except as provided in paragraphs (c)(2)(i) and (ii) of this 
section, if a coke oven door leak is observed at any time during the 
coking cycle, the owner or operator shall take corrective action and 
stop the leak within 15 minutes from the time the leak is first 
observed. No additional leaks are allowed from doors on that oven for 
the remainder of that oven's coking cycle.
    (i) Except as provided in paragraph (c)(2)(ii) of this section, the 
owner or operator may take corrective action and stop the leak within 
45 minutes (instead of 15 minutes) from the time the leak is first 
observed for a maximum of two times per battery in any semiannual 
reporting period.
    (ii) If a worker must enter a cokeside shed to stop a leaking door 
under the cokeside shed, the owner or operator shall take corrective 
action and stop the door leak within 45 minutes (instead of 15 minutes) 
from the time the leak is first observed. The evacuation system and 
control device for the cokeside shed must be operated at all times 
there is a leaking door under the cokeside shed.
    (d) The owner or operator of a new nonrecovery coke oven battery 
shall meet the emission limitations and work practice standards in 
paragraphs (d)(1) through (4) of this section.
    (1) The owner or operator shall not discharge or cause to be 
discharged to the atmosphere from charging operations any fugitive 
emissions that exhibit an opacity greater than 20 percent, as 
determined by the procedures in Sec.  63.309(j).
    (2) The owner or operator shall not discharge or cause to be 
discharged to the atmosphere any emissions of particulate matter (PM) 
from a charging emissions control device that exceed 0.0081 pounds per 
ton (lbs/ton) of dry coal charged, as determined by the procedures in 
Sec.  63.309(k).
    (3) The owner or operator shall observe the exhaust stack of each 
charging emissions control device at least once each day of operation 
during charging to determine if visible emissions are present and shall 
record the results of each daily observation or the reason why 
conditions did not permit a daily observation. If any visible emissions 
are observed, the owner or operator must:
    (i) Take corrective action to eliminate the presence of visible 
emissions;
    (ii) Record the cause of the problem creating the visible emissions 
and the corrective action taken;
    (iii) Conduct visible emission observations according to the 
procedures in Sec.  63.309(m) within 24 hours after detecting the 
visible emissions; and
    (iv) Report any 6-minute average, as determined according to the 
procedures in Sec.  63.309(m), that exceeds 10 percent opacity as a 
deviation in the semiannual compliance report required by Sec.  
63.311(d).
    (4) The owner or operator shall develop and implement written 
procedures for adjusting the oven uptake damper to maximize oven draft 
during charging and for monitoring the oven damper setting during each 
charge to ensure that the damper is fully open.

0
6. Section 63.309 is amended by adding new paragraphs (j) through (m) 
to read as follows:


Sec.  63.309  Performance tests and procedures.

* * * * *
    (j) The owner or operator of a new nonrecovery coke oven battery 
shall conduct a performance test once each week to demonstrate 
compliance with the opacity limit in Sec.  63.303(d)(1). The owner or 
operator shall conduct each performance test according to the 
procedures and requirements in paragraphs (j)(1) through (3) of this 
section.
    (1) Using a certified observer, determine the average opacity of 
five consecutive charges per week for each charging emissions capture 
system if charges can be observed according to the requirements of 
Method 9 (40 CFR part 60, appendix A), except as specified in 
paragraphs (j)(1)(i) and (ii) of this section.
    (i) Instead of the procedures in section 2.4 of Method 9 (40 CFR 
part 60, appendix A), record observations to the nearest 5 percent at 
15-second intervals for at least five consecutive charges.
    (ii) Instead of the procedures in section 2.5 of Method 9 (40 CFR 
part 60, appendix A), determine and record the highest 3-minute average 
opacity for each charge from the consecutive observations recorded at 
15-second intervals.
    (2) Opacity observations are to start when the door is removed for 
charging and end when the door is replaced.
    (3) Using the observations recorded from each performance test, the 
certified observer shall compute and record the average of the highest 
3-minute averages for five consecutive charges.
    (k) The owner or operator of a new nonrecovery coke oven battery 
shall conduct a performance test to demonstrate initial compliance with 
the emission limitations for a charging

[[Page 20014]]

emissions control device in Sec.  63.303(d)(2) within 180 days of the 
compliance date that is specified for the affected source in Sec.  
63.300(a)(4) and report the results in the notification of compliance 
status. The owner or operator shall prepare a site-specific test plan 
according to the requirements in Sec.  63.7(c) and shall conduct each 
performance test according to the requirements in Sec.  63.7(e)(1) and 
paragraphs (k)(1) through (4) of this section.
    (1) Determine the concentration of PM according to the following 
test methods in appendix A to 40 CFR part 60.
    (i) Method 1 to select sampling port locations and the number of 
traverse points. Sampling sites must be located at the outlet of the 
control device and prior to any releases to the atmosphere.
    (ii) Method 2, 2F, or 2G to determine the volumetric flow rate of 
the stack gas.
    (iii) Method 3, 3A, or 3B to determine the dry molecular weight of 
the stack gas. You may also use as an alternative to Method 3B, the 
manual method for measuring the oxygen, carbon dioxide, and carbon 
monoxide content of exhaust gas, ANSI/ASME PTC 19.10-1981, ``Flue and 
Exhaust Gas Analyses'' (incorporated by reference, see Sec.  63.14).
    (iv) Method 4 to determine the moisture content of the stack gas.
    (v) Method 5 or 5D, as applicable, to determine the concentration 
of front half PM in the stack gas.
    (2) During each PM test run, sample only during periods of actual 
charging when the capture system fan and control device are engaged. 
Collect a minimum sample volume of 30 dry standard cubic feet (dscf) 
during each test run. Three valid test runs are needed to comprise a 
performance test. Each run must start at the beginning of a charge and 
finish at the end of a charge (i.e., sample for an integral number of 
charges).
    (3) Determine and record the total combined weight of tons of dry 
coal charged during the duration of each test run.
    (4) Compute the process-weighted mass emissions (Ep) for 
each test run using Equation 1 of this section as follows:
[GRAPHIC] [TIFF OMITTED] TR15AP05.000

Where:

Ep = Process weighted mass emissions of PM, lb/ton;
C = Concentration of PM, grains per dry standard cubic foot (gr/dscf);
Q = Volumetric flow rate of stack gas, dscf/hr;
T = Total time during a run that a sample is withdrawn from the stack 
during charging, hr;
P = Total amount of dry coal charged during the test run, tons; and
K = Conversion factor, 7,000 grains per pound (gr/lb).

    (l) The owner or operator of a new nonrecovery coke oven battery 
shall conduct subsequent performance tests for each charging emissions 
control device subject to the PM emissions limit in Sec.  63.303(d)(2) 
at least once during each term of their title V operating permit.
    (m) Visible emission observations of a charging emissions control 
device required by Sec.  63.303(d)(3)(iii) must be performed by a 
certified observer according to Method 9 (40 CFR part 60, appendix A) 
for one 6-minute period.

0
7. Section 63.310 is amended by adding new paragraph (j) to read as 
follows:


Sec.  63.310  Requirements for startups, shutdowns, and malfunctions.

* * * * *
    (j) The owner or operator of a nonrecovery coke oven battery 
subject to the work practice standards for door leaks in Sec.  
63.303(c) shall include the information specified in paragraphs (j)(1) 
and (2) of this section in the startup, shutdown, and malfunction plan.
    (1) Identification of potential malfunctions that will cause a door 
to leak, preventative maintenance procedures to minimize their 
occurrence, and corrective action procedures to stop the door leak.
    (2) Identification of potential malfunctions that affect charging 
emissions, preventative maintenance procedures to minimize their 
occurrence, and corrective action procedures.

0
8. Section 63.311 is amended as follows:
0
a. Revising paragraph (b)(1) and adding new paragraphs (b)(3) through 
(7);
0
b. Revising paragraph (c)(1) and adding new paragraph (c)(3);
0
c. Revising paragraphs (d)(1) through (3) and adding new paragraphs 
(d)(4) through (9); and
0
d. Revising paragraphs (f)(1)(i) and (ii) and adding new paragraphs 
(f)(1)(iv) through (ix).


Sec.  63.311  Reporting and recordkeeping requirements.

* * * * *
    (b) * * *
    (1) Statement signed by the owner or operator, certifying that a 
bypass/bleeder stack flare system or an approved alternative control 
device or system has been installed as required in Sec.  63.307.
    (2) * * *
    (3) Statement, signed by the owner or operator, certifying that all 
work practice standards for charging operations have been met as 
required in Sec.  63.303(b)(3).
    (4) Statement, signed by the owner or operator, certifying that all 
work practice standards for door leaks have been met as required in 
Sec.  63.303(c).
    (5) Statement, signed by the owner or operator, certifying that the 
information on potential malfunctions has been added to the startup, 
shutdown and malfunction plan as required in Sec.  63.310(j).
    (6) Statement, signed by the owner or operator, that all applicable 
emission limitations in Sec.  63.303(d)(1) and (2) for a new 
nonrecovery coke oven battery have been met. The owner or operator 
shall also include the results of the PM performance test required in 
Sec.  63.309(k).
    (7) Statement, signed by the owner or operator, certifying that all 
work practice standards in Sec.  63.303(d)(3) and (4) for a new 
nonrecovery coke oven battery have been met.
    (c) * * *
    (1) Intention to construct a new coke oven battery (including 
reconstruction of an existing coke oven battery and construction of a 
greenfield coke oven battery), a brownfield coke oven battery, or a 
padup rebuild coke oven battery, including the anticipated date of 
startup.
    (2) * * *
    (3) Intention to conduct a PM performance test for a new 
nonrecovery coke oven battery subject to the requirements in Sec.  
63.303(d)(2). The owner or operator shall provide written notification 
according to the requirements in Sec.  63.7(b).
    (d) * * *
    (1) Certification, signed by the owner or operator, that no coke 
oven gas was vented, except through the bypass/bleeder stack flare 
system of a by-product coke oven battery during the reporting period or 
that a venting report has been submitted according to the requirements 
in paragraph (e) of this section.
    (2) Certification, signed by the owner or operator, that a startup, 
shutdown, or malfunction event did not occur for a coke oven battery 
during the reporting period or that a startup, shutdown, and 
malfunction event did occur and a report was submitted according to the 
requirements in Sec.  63.310(e).
    (3) Certification, signed by the owner or operator, that work 
practices were implemented if applicable under Sec.  63.306.

[[Page 20015]]

    (4) Certification, signed by the owner or operator, that all work 
practices for nonrecovery coke oven batteries were implemented as 
required in Sec.  63.303(b)(3).
    (5) Certification, signed by the owner or operator, that all coke 
oven door leaks on a nonrecovery battery were stopped according to the 
requirements in Sec.  63.303(c)(2) and (3). If a coke oven door leak 
was not stopped according to the requirements in Sec.  63.303(c)(2) and 
(3), or if the door leak occurred again during the coking cycle, the 
owner or operator must report the information in paragraphs (d)(5)(i) 
through (iii) of this section.
    (i) The oven number of each coke oven door for which a leak was not 
stopped according to the requirements in Sec.  63.303(c)(2) and (3) or 
for a door leak that occurred again during the coking cycle.
    (ii) The total duration of the leak from the time the leak was 
first observed.
    (iii) The cause of the leak (including unknown cause, if 
applicable) and the corrective action taken to stop the leak.
    (6) Certification, signed by the owner or operator, that the 
opacity of emissions from charging operations for a new nonrecovery 
coke oven battery did not exceed 20 percent. If the opacity limit in 
Sec.  63.303(d)(1) was exceeded, the owner or operator must report the 
number, duration, and cause of the deviation (including unknown cause, 
if applicable), and the corrective action taken.
    (7) Results of any PM performance test for a charging emissions 
control device for a new nonrecovery coke oven battery conducted during 
the reporting period as required in Sec.  63.309(l).
    (8) Certification, signed by the owner or operator, that all work 
practices for a charging emissions control device for a new nonrecovery 
coke oven battery were implemented as required in Sec.  63.303(d)(3). 
If a Method 9 (40 CFR part 60, appendix A) visible emissions 
observation exceeds 10 percent, the owner or operator must report the 
duration and cause of the deviation (including unknown cause, if 
applicable), and the corrective action taken.
    (9) Certification, signed by the owner or operator, that all work 
practices for oven dampers on a new nonrecovery coke oven battery were 
implemented as required in Sec.  63.303(d)(4).
* * * * *
    (f) * * *
    (1) * * *
    (i) Records of daily pressure monitoring, if applicable according 
to Sec.  63.303(a)(1)(ii) or Sec.  63.303(b)(1)(ii).
    (ii) Records demonstrating the performance of work practice 
requirements according to Sec.  63.306(b)(7). This requirement applies 
to nonrecovery coke oven batteries subject to the work practice 
requirements in Sec.  63.303(a)(2) or Sec.  63.303(b)(3).
    (iii) * * *
    (iv) Records to demonstrate compliance with the work practice 
requirement for door leaks in Sec.  63.303(c). These records must 
include the oven number of each leaking door, total duration of the 
leak from the time the leak was first observed, the cause of the leak 
(including unknown cause, if applicable), the corrective action taken, 
and the amount of time taken to stop the leak from the time the leak 
was first observed.
    (v) Records to demonstrate compliance with the work practice 
requirements for oven uptake damper monitoring and adjustments in Sec.  
63.303(c)(1)(iv).
    (vi) Records of weekly performance tests to demonstrate compliance 
with the opacity limit for charging operations in Sec.  63.303(d)(1). 
These records must include calculations of the highest 3-minute 
averages for each charge, the average opacity of five charges, and, if 
applicable, records demonstrating why five consecutive charges were not 
observed (e.g., the battery was charged only at night).
    (vii) Records of all PM performance tests for a charging emissions 
control device to demonstrate compliance with the limit in Sec.  
63.303(d)(2).
    (viii) Records of all daily visible emission observations for a 
charging emission control device to demonstrate compliance with the 
requirements limit in Sec.  63.303(d)(3).
    (ix) Records to demonstrate compliance with the work practice 
requirements for oven uptake damper monitoring and adjustments in Sec.  
63.303(d)(4).
* * * * *
[FR Doc. 05-6942 Filed 4-14-05; 8:45 am]
BILLING CODE 6560-50-P